An emerging generation of reactor designs has reinvigorated the nuclear power sector. These reactors, intended to limit cost and scale, represent a dramatical departure from the previous three nuclear power generations. Demonstration reactors and prototypes are developing quickly, and many developers anticipate operation beginning during the 2030s. The fuels, coolants, and other features of these designs will affect all stages of the nuclear fuel cycle, including at the back-end, after fuel is removed from a reactor. Meanwhile, developments in long-term nuclear waste management and final disposal of spent nuclear fuel (SNF) are continuing. Moreover, existing concerns about the hazards of SNF make waste management a sensitive topic, and these new reactors will compound challenges in waste management with new and different SNF volumes and compositions. As disposal plans and these new reactors develop in parallel, stakeholders should seize the opportunity to address existing and emerging SNF management concerns. This working paper explores SNF management for emerging reactors as an essential part of the preparation for their deployment. Starting with an examination of the new fuels and other design features, as well as comparative approximations of SNF output, the paper assesses major technical, capacity, and political challenges for emerging reactors at the back-end of the fuel cycle. The paper focuses in particular on challenges in spent fuel treatment and disposal, reprocessing and recycling, and the implementation and inclusion of international nuclear safeguards.
Nuclear power generation has been largely unchanged over the past 50 years, with a reliance primarily on large, water-cooled reactors executing once-through fuel cycles. The new “Generation IV” reactors propose to alter how nuclear power is conceptualized. Many of the almost 140 advanced reactor designs worldwide build on existing experimental designs that have yet to be widely commercially deployed or demonstrated, including liquid and metal fuels and non-light-water cooling.1Third Way, “The Global Race for Advanced Nuclear,” John Milko and Todd Allen, (2017), accessed November 6, 2020, https://www.thirdway.org/infographic/the-global-race-for-advanced-nuclear These features are to be incorporated into what aim to be safe, small, cost-effective reactors potentially with built-in nonproliferation benefits. As disruptive technologies, they bring with them a host of new challenges. Emerging reactors will change the composition and volume of fuel, particularly in the back-end, or post-reactor, and some will operate by recycling spent — or used — nuclear fuel (SNF) in closed fuel cycles.2See Appendix D: Glossary of Terms for relevant abbreviations. The differences between emerging and conventional reactors will require consideration throughout the emerging reactor development process. These changes to the conventional fuel cycle will compound existing challenges in long-term SNF management and safeguards implementation while providing opportunities for reassessing practices.
Waste from emerging reactors will add another layer of complexity to SNF management. SNF management has always been a contentious topic as a result of the material’s hazards and potential weaponization risk. As of the end of 2016, around 400,000 metric tons of heavy metal of SNF from commercial reactors and research and test reactors were in storage at reactor sites or other facilities.3Stimson Center, “Spent Nuclear Fuel Storage and Disposal, Stimson Center,” Trinh Le, (2020), accessed October 9, 2020, https://www.stimson.org/2020/spent-nuclear-fuel-storage-and-disposal/; global SNF figure based on national reports submitted in accordance with the IAEA’s Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management: IAEA, “Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management,” 2021, accessed January 26, 2021, https://www.iaea.org/topics/nuclear-safety-conventions/joint-convention-safety-spent-fuel-management-and-safety-radioactive-waste Inventory has grown faster than storage facility capacity around the world, requiring countries facing capacity challenges to retroactively expand interim storage facilities. After nearly 50 years of global research and the accumulation of hundreds of thousands of metric tons of SNF in storage, countries are slowly advancing towards final disposal of high-level waste (HLW) and SNF in deep geological repositories (DGRs), with the world’s first such repository finally under construction in Finland.4Stimson Center, “Spent Nuclear Fuel Storage and Disposal.” As DGR development coincides with emerging reactors’ disruption to the nuclear power sector, reactor developers and policy makers have an opportunity to address SNF management, disposal, and back-end international safeguards challenges. Emerging reactors present a variety of potential benefits and challenges including in security and cost, so while this paper is not exhaustive, it aims to highlight several of the major challenges that could exacerbate existing back-end issues. This paper examines how SNF from advanced fission reactors differs from the existing fleet and provides comparative SNF quantity approximations. It explores potential technical and policy challenges in SNF disposal, reprocessing and recycling, and safeguards applications to demonstrate the importance of considering back-end management during the planning for emerging reactor deployment.
Status of the Current Fleet
First-generation commercial nuclear power production, adapted from 1940s naval reactor programs, began at a small scale in the 1950s.5American Academy of Arts & Sciences, “Nuclear Reactors: Generation to Generation,” Stephen M. Goldberg and Robert Rosner (2011), 3, accessed November 6, 2020, https://www.amacad.org/publication/nuclear-reactors-generation-generation; U.S. Department of Energy, The History of Nuclear Energy, DOE/NE-0088, 8, accessed November 6, 2020, https://www.energy.gov/sites/prod/files/The%20History%20of%20Nuclear%20Energy_0.pdf Second-generation reactors include the reactor types we see today, with 40-year operating lifetimes and larger outputs.6American Academy of Arts & Sciences, “Nuclear Reactors,” 4. Generations III and III+, initiated in the 1990s and continuing today with such designs as the Advanced Boiling Water Reactor and Advanced CANDU Reactor, build on existing designs and aim to expand energy capacity.7American Academy of Arts & Sciences, “Nuclear Reactors,” 7, 8. Together, these generations have developed the earliest practical designs for nuclear power and comprise the overwhelming bulk of global nuclear power today; within their operating lifetimes, some have seen nearly the entire expansion of the nuclear power sector. Throughout these generations, nuclear power has been controversial, in part because of accidents including Chernobyl and Fukushima, and the unending question of what to do with nuclear waste. Generation IV reactors, not yet widely commercialized, aim to reinvigorate nuclear power by increasing safety, improving cost efficiency, and addressing spent fuel concerns.8Generation IV International Forum, “Gen IV Reactor Design,” September 26, 2013, accessed January 14, 2021, https://www.gen-4.org/gif/jcms/c_40275/faq To do so, these reactors may employ materials, structures, or techniques distinct from those of the current fleet. See Figure 1 for an overview of the nuclear power generations.
Fig. 1: Nuclear Power Reactor Generations I Through IV.9American Academy of Arts & Sciences, “Nuclear Reactors,” 4, 6, 7, 14.
Almost all current commercial reactors are thermal spectrum reactors, meaning they slow neutrons to promote fission reactions. In addition to the typical uranium fuel comprising fertile U-238 material and around five percent fissile U-235, thermal reactors can use a mixed plutonium and uranium oxide (MOX) fuel resulting from the reprocessing of fissile material from spent fuel.10IAEA, “The Nuclear Fuel Cycle,” 11, 21, accessed January 29, 2021, https://www.iaea.org/sites/default/files/19/02/the-nuclear-fuel-cycle.pdf Alternative thorium-uranium fuel cycles have also been an experimental — but not commercialized — option to use thermal reactors as “breeders,” which produce more fissile material than they burn.11Thorium cycles have not yet been widely implemented. IAEA, Thorium Fuel Cycle – Potential Benefits and Challenges, IAEA-TECDOC-1450, Vienna, (2005), 4, 8, accessed November 26, 2020, https://www.iaea.org/publications/7192/thorium-fuel-cycle-potential-benefits-and-challenges There are currently a small number of fast reactors operating, which, unlike thermal reactors, do not slow neutrons to control fission and use non-water coolants to keep neutrons at higher — “fast” — energies.12Nick Touran, “What Is a Fast Reactor?,” What Is Nuclear?, last modified September 2009, accessed October 29, 2020, https://whatisnuclear.com/fast-reactor.html ; Frank von Hippel, “Overview: The Rise and Fall of Plutonium Breeder Reactors,” in Fast Breeder Reactor Programs: History and Status, IPFM Research Report #8, (International Panel on Fissile Materials, February 2010), 4, http://fissilematerials.org/library/rr08.pdf They can also be breeders within more conventional uranium and plutonium fuel cycles, and can fission minor actinides (MAs).13Nuclear Engineering Division, “Reactors Designed by Argonne National Laboratory,” accessed November 6, 2020, https://www.ne.anl.gov/About/reactors/frt.shtml
The following reactor types do not include all existing commercial reactors but focus on the most widespread and influential operating designs to provide a starting point for comparison with emerging reactors.Variants on typical reactors, such as the Russian VVER-type pressurized water reactor (PWR), and other designs including the United Kingdom’s advanced gas-cooled reactor (AGR) are not explored in depth here for simplicity. But these and other reactors have provided foundations for innovative emerging reactors. For example, fast breeder reactors (FBRs) have been under consideration since the earliest days of nuclear power, but enthusiasm waned after the discovery that uranium was more abundant than expected, that nuclear expansion was not as urgent, and that FBRs were not cost effective at that time.14Von Hippel, “Overview: the Rise and Fall of Plutonium Breeder Reactors,” 1, 5. Now, with renewed interest in unconventional reactors, FBRs are seeing a resurgence. Fast reactors will be discussed in more depth in the section on “Emerging Reactor Classes and Their SNF.”
For a comparative chart of the reactors discussed, see Figure 2 at the end of the section on “Emerging Reactor Classes and Their SNF.” More detail about the approximations made for SNF mass and volume outputs can be found in Appendices A and B, respectively.
Light Water Reactor (LWR)
Of the current global fleet, 377 of 443 reactors (85 percent) are thermal light water reactors.15IAEA, “Power Reactor Information System,” accessed January 27, 2021, https://pris.iaea.org/PRIS/home.aspx These reactors use light water as coolant around solid fuel rods in square assemblies, typically of low enriched (less than five percent) uranium dioxide.16World Nuclear Association, Nuclear Power Reactor Characteristics, (2018), 1, accessed November 6, 2020, http://www.world-nuclear.org/getmedia/80f869be-32c8-46e7-802d-eb4452939ec5/Pocket-Guide-Reactors.pdf.aspx Some LWRs also use MOX fuel. Of that 85 percent, the most prevalent are PWRs.17Here, PWRs includes all models — for example the Westinghouse design and the Russian VVER, which first operated in 1964. Rosatom, “Modern Reactors of Russian Design,” accessed October 30, 2020, https://rosatom.ru/en/rosatom-group/engineering-and-construction/modern-reactors-of-russian-design/ These reactors have a typical upper burnup of 50 gigawatt-days per metric ton (GWd/MT) and a power density of around 100 megawatts per cubic meter (MW/m3).18IAEA, “Advanced Reactors Information System,” last modified August 2012, accessed November 6, 2020, https://aris.iaea.org/; Jacopo Buongiorno, “PWR Description,” class lecture, Engineering of Nuclear Systems, MIT, Boston, MA, 2010, 11, https://ocw.mit.edu/courses/nuclear-engineering/22-06-engineering-of-nuclear-systems-fall-2010/lectures-and-readings/MIT22_06F10_lec06a.pdf Their thermal efficiency is around 33 percent.19U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, Brian K. Castle et al., (September 2012), 6. https://inldigitallibrary.inl.gov/sites/sti/sti/5554578.pdf LWRs produce about 54.55 kg SNF per unit of energy output, and a fuel volume per gigawatt-year of electrical power (GWye) of 6.53e-3 m3.20See Appendix A for approximation of mass and Appendix B for approximation of volume. Spent fuel rods as well as the rods’ cladding material comprise the most highly radioactive material in storage.
The other common LWR is the boiling water reactor (BWR), which uses boiling water as the coolant around uranium oxide fuel rods. The burnup and thermal efficiency, like that of PWRs, typically reaches 50 GWd/MT and 33 percent respectively, while power density is around 55 MW/m3 on average.21U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 7; IAEA, “Advanced Reactors Information System.” SNF mass per unit of energy output is the same as for PWRs, while volume per GWye is around 10.72e-3 m3.22See Appendix A for mass and Appendix B for volume. Like PWRs, spent fuel rods and cladding are the most radioactive waste products.
Pressurized Heavy Water Reactor (PHWR)
The thermal spectrum PHWR, most notably the Canadian Deuterium Uranium (CANDU) model, is the third most prevalent type of the current fleet with 48 operational reactors.23IAEA, “Power Reactor Information System.” The CANDU reactor uses heavy water, or deuterium oxide, as the coolant and natural uranium oxide — as opposed to the more typical enriched — as the fuel.24U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 7. The thermal efficiency is around 30 percent and burnup is low at 7 GWd/MT.25U.S Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 7, 8. Power density is approximately 10 MW/m3.26IAEA, “Other Designs of Nuclear Power Stations,” Nuclear Graphite Knowledge Base, last modified 2020, accessed October 9, 2020, https://nucleus.iaea.org/sites/graphiteknowledgebase/wiki/Guide_to_Graphite/Other%20Designs%20of%20Nuclear%20Power%20Stations.aspx SNF arises in the form of spent fuel elements and cladding in cylindrical “bundles,” which are smaller and configured differently from LWR fuel assemblies. CANDU SNF mass per unit of energy output can be eight times that of LWRs, as a result of natural uranium and comparatively low burnup and thermal efficiency.27See Appendix A. Volume is also higher at 16.54 m3/GWye.28See Appendix B.
Together, PWRs, BWRs, and PHWRs/CANDU represent about 93 percent of the present operational fleet.29IAEA, “Power Reactor Information System,” accessed January 14, 2021, https://pris.iaea.org/PRIS/home.aspx Many emerging reactor designs are vastly different from these reactors in fuel form, coolant, or other features, which will result in new amounts and types of spent fuel outputs through operation. Those wastes are described in more detail below.
Emerging Reactor Classes and Their SNF
The first three generations of nuclear power have largely built on consistent designs with previous operating experience. Even as Generation III+ designs are still developing, an emerging new generation of reactors departs significantly from the previous three generations’ legacy. While this new set of reactors is called “Generation IV,” the impacts it brings to the nuclear fuel cycle points to a second wave in nuclear power conceptualization, different from the approaches that supported the previous three generations.
While there have been historical precedents for FBR types and others discussed in this section, what makes the development of Generation IV designs notable is the more mainstream effort for commercial deployment.
These reactors are steadily moving toward deployment, with two operational sodium-cooled FBRs (in Russia), configured to breed more fissile material than they burn, bridging the gap between the current operational fleet and new nuclear.30The two Russian FBRs commenced commercial operation in 1981 and 2016. IAEA, “Power Reactor Information System.” At the time of writing, the third FBR, in China, is not producing power, but is operable.31IAEA, “Power Reactor Information System.” Previously, FBRs have only produced electricity in states identified as nuclear-weapon States under the Treaty on the Non-Proliferation of Nuclear Weapons (NPT).32Article IX (3) defines nuclear-weapon State as one that manufactured and exploded a nuclear weapon or nuclear device by 1 January 1967. Five States are therefore recognized as nuclear-weapon States: China, France, Russia, the United Kingdom, and the United States. United Nations Office for Disarmament Affairs, “Treaty on the Non-Proliferation of Nuclear Weapons (NPT),” accessed January 19, 2021, https://www.un.org/disarmament/wmd/nuclear/npt/text Japan’s Monju, operational only in 1995, did not commercially produce electricity.33IAEA, “Power Reactor Information System.” There is renewed discussion around FBRs now that India, a nuclear-weapon-possessor outside the NPT, is constructing an FBR to begin closing its fuel cycle. This construction is raising concerns about the difficulty of verifying peaceful use in a country with limited International Atomic Energy Agency (IAEA) safeguards obligations.34M.V. Ramana, “India and Fast Breeder Reactors,” in Fast Breeder Reactor Programs: History and Status, IPFM Research Report #8 (International Panel on Fissile Materials: February 2010), 40, http://fissilematerials.org/library/rr08.pdf While there have been historical precedents for FBR types and others discussed in this section, what makes the development of Generation IV designs notable is the more mainstream effort for commercial deployment.
The Generation IV International Forum, a cohort of 14 countries cooperating in advanced nuclear energy research and development, has identified six next-generation reactor classes with potential for future deployment: very high temperature reactors (VHTRs; thermal), gas-cooled fast reactors (GFRs; fast), sodium-cooled fast reactors (SFRs; fast), lead-cooled fast reactors (LFRs; fast), molten salt reactors (MSRs; thermal/fast), and supercritical water-cooled reactors (SCWRs; thermal/fast).35Generation IV International Forum, “Technology Systems,” last modified 2019, accessed November 6, 2020, https://www.gen-4.org/gif/jcms/c_40486/technology-systems These innovative reactor classes, three of which the U.S. Department of Energy lists as “3 Advanced Reactor Systems to Watch by 2030,” will be explored in more depth below.36U.S. Department of Energy, “3 Advanced Reactor Systems to Watch by 2030,” last modified March 7, 2018, accessed November 6, 2020, https://www.energy.gov/ne/articles/3-advanced-reactor-systems-watch-2030
The nuclear industry is also currently pursuing small modular reactors. These reactors, which can use a variety of coolant and fuel types, are characterized by their electric output of 300 megawatts electric (MWe) or less and their modular, factory-produced components, which are intended to be quickly and more cost-effectively constructed, with more siting flexibility. The Generation IV classes have the potential to more significantly impact SNF composition and quantity than do scaled-down small water-cooled reactors, so this paper focuses on the waste impacts of the more unconventional Generation IV identified reactors of variable size.
A 2019 IAEA preliminary study of waste types from innovative reactors indicated that this kind of work is gradually becoming more prevalent.37IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, Nuclear Energy Series NW-T-1.7, (2019), https://www-pub.iaea.org/MTCD/Publications/PDF/PUB1822_web.pdf Informed by that study and other publications, this paper studies operational SNF volume and composition in comparison to existing commercial LWRs as the most prevalent power reactors. A comprehensive prediction of total emerging reactor SNF arising at the national level is beyond the scope of this paper because the number and location of planned emerging reactors is still unknown. However, it is possible to approximate the differences on a unit-by-unit basis. New reactor classes will likely have a significant impact on SNF generation and proliferation control as a result of departures from existing reactors in SNF volume and composition, and the subsequent potential reprocessing and recycling.
Along with SNF, these reactors produce different amounts or types of lower-level radioactive wastes from their operation than do conventional reactors.38The IAEA provides a document classifying radioactive waste, but based on differing national policies, countries may or may not consider SNF to be a form of HLW. IAEA, Classification of Radioactive Waste, IAEA Safety Standards Series No. GSG-1, (2009), https://www.iaea.org/publications/8154/classification-of-radioactive-waste In fact, options to use alternative fuel cycles to reduce SNF inventories could increase the amount of lower level wastes to be disposed of as compared to conventional once-through cycles.39Nuclear Waste Management Organization (Canada), Watching Brief on Advanced Fuel Cycles: 2019 Update, (Ontario: NWMO, 2020), 13, https://www.nwmo.ca/~/media/Site/Reports/2020/03/18/18/15/Watching-brief-on-advanced-fuel-cycles–2019-update–EN.ashx?la=en As with SNF, the long-term management of lower-level radioactive waste is the topic of ongoing and contentious discussion. Low- and intermediate-level waste (LILW) can be also be disposed of underground in repositories, such as in the Wolsung LILW Disposal Center in the Republic of Korea.40For more information on this facility, see Cindy Vestergaard and Trinh Le, “Exploring the Wolsung LILW Disposal Center in South Korea,” Stimson Center, last modified August 7, 2019, accessed October 22, 2020, https://www.stimson.org/2019/exploring-wolsung-lilw-disposal-center-south-korea/ Several examples of lower-level wastes are shared throughout the paper to illustrate the larger changes to back-end waste management, but their management will not be explored in depth. More study into emerging reactors’ lower-level waste volumes and compositions should be pursued.
Very High Temperature Reactor (VHTR)
VHTRs are helium-cooled thermal spectrum reactors based on the high temperature gas-cooled reactor (HTGR) design.41Generation IV International Forum, “Very-High-Temperature Reactor (VHTR),” last modified 2019, accessed November 6, 2020, https://www.gen-4.org/gif/jcms/c_42153/very-high-temperature-reactor-vhtr They can potentially operate at temperatures ranging from 700 to 950˚C, two to three times higher than LWRs.42IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 6; LWR outlet temperatures from Buongiorno, “PWR Description,” 11, and Jacopo Buongiorno, “BWR Description,” class lecture, Engineering of Nuclear Systems, MIT, Boston, MA, 2010, 3, https://ocw.mit.edu/courses/nuclear-engineering/22-06-engineering-of-nuclear-systems-fall-2010/lectures-and-readings/MIT22_06F10_lec06b.pdf The thermal efficiency is around 45 percent.43Institut de Radioprotection et de Sûreté Nucléaire (IRSN) (France), Review of Generation IV Nuclear Energy Systems, (Fontenay-aux-Roses: IRSN, 27 April, 2015), 51, https://www.irsn.fr/EN/newsroom/News/Documents/IRSN_Report-GenIV_04-2015.pdf VHTRs have significantly lower power densities than LWRs at around 4 to 10 MW/m3.44IRSN, Review of Generation IV Nuclear Energy Systems, 51; They can have burnups from 90 to 120 GWd/MT as compared to an LWR’s higher burnup of 50 GWd/MT and a CANDU burnup of 7 GWd/MT, and they are fueled with tristructural isotropic (TRISO) particle fuel in prismatic blocks or pebbles.45U.S. Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), David L. Moses, (2010), 4, 6, https://www.osti.gov/biblio/1027406/ ; Yuji Fukaya and Tetsuo Nishihara, “Reduction on High Level Radioactive Waste Volume and Geological Repository Footprint With High Burn-up and High Thermal Efficiency of HTGR,” Nuclear Engineering and Design 307 (October 2016): 190, accessed November 16, http://dx.doi.org/10.1016/j.nucengdes.2016.07.009 TRISO fuel is typically part of a once-through fuel cycle but can be used in alternative MOX, plutonium, or other fuel cycles.46Columbia University Center on Global Energy Policy, A Comparison of Advanced Nuclear Technologies, Andrew C. Kadak (2017), 53, https://energypolicy.columbia.edu/sites/default/files/A%20Comparison%20of%20Nuclear%20Technologies%20033017.pdf; Generation IV International Forum, “Very-High-Temperature Reactor (VHTR).” The pebble type can be refueled online, with individual pebbles recirculated until they reach their burnup limits, which occurs after about three years.47IAEA, High Temperature Gas Cooled Reactor Fuels and Materials, (AEA TECODC (CD-ROM) No. 1645 (2010), 61, https://www.iaea.org/publications/8270/high-temperature-gas-cooled-reactor-fuels-and-materials The prismatic type has a more conventional refueling period of approximately every 18 months.48Based on existing designs in U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), B-4. The spent TRISO includes irradiated fuel particles in the TRISO coatings. Unlike conventional solid fuels, TRISO fuel does not produce irradiated cladding that is separated from the fuel after removal.49IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 19. TRISO fuel consists of fuel particles suspended in a matrix, surrounded by coating layers that act as the cladding.50U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), 5. This coating is more difficult to separate from the fuel than conventional cladding is and can provide a containment barrier, making it unlikely that the coating would be separated from the fuel prior to disposal.51IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 106–107; U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), 17. VHTRs do produce tritium waste at a higher rate than LWRs through graphite moderation; this waste can be managed through storage as it decays or through immobilization for disposal.52IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 27. For more information on tritium management see IAEA, Management of Waste Containing Tritium and Carbon-14, Technical Reports Series No. 421 (2004), https://www.iaea.org/publications/6634/management-of-waste-containing-tritium-and-carbon-14 SNF mass per unit of energy output is around 31 to 38 percent that of an LWR, but SNF volume will be higher than that of an LWR, potentially reaching around 24.4e-3 m3/GWye.53See Appendices A and B for SNF output approximations; IAEA, “Advanced Reactors Information System”; possibly with the exception of a thorium-cycle VHTR, in which volume could be lower per megawatt than a PWR, according to IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 19; Columbia University Center on Global Energy Policy, A Comparison of Advanced Nuclear Technologies, 53.
Gas-cooled Fast Reactor (GFR)
Like VHTRs, GFRs are cooled with helium gas and operate at high temperature. They will likely use solid fuel rods of uranium or plutonium carbide, nitride, or oxide.54Generation IV International Forum, GIF R&D Outlook for Generation IV Nuclear Energy Systems: 2018 Update (2019), 43, 44, 45, https://www.gen-4.org/gif/jcms/c_108744/gif-r-d-outlook-for-generation-iv-nuclear-energy-systems-2018-update?details=true Average burnup is approximately 120 GWd/MT and thermal efficiency ranges from 33 to 53 percent.55Burnup and thermal efficiency from IAEA, “Advanced Reactors Information System.” Power density can range from 53 to 100 MW/m3.56IAEA, “Advanced Reactors Information System.” Their spent fuel mass per unit of energy output is about 33 percent that of an LWR.57See Appendix A. GFRs produce a smaller SNF volume than LWRs, at 0.774e-3 m3/GWye.58See Appendix B. They can operate in closed cycles, where SNF is reprocessed to extract fission products and turn the remaining material into usable fuel.59Generation IV International Forum, “Gas-Cooled Fast Reactor,” last modified 2019, accessed November 6, 2020, https://www.gen-4.org/gif/jcms/c_9357/gfr They can be configured as breeder reactors.60Generation IV International Forum, GIF R&D Outlook for Generation IV Nuclear Energy Systems: 2018 Update, 23. The GFR’s notable waste beyond SNF is ceramic LILW from its ceramic-covered carbide or nitride fuel which requires disposal.61IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 23, 36.
Sodium-cooled Fast Reactor (SFR)
SFRs use liquid sodium metal coolant and solid fuel rods and have the most experimental experience of the Generation IV reactors.62Generation IV International Forum, “Sodium-Cooled Fast Reactor (SFR),” last modified 2019, accessed November 6, 2020, https://www.gen-4.org/gif/jcms/c_42152/sodium-cooled-fast-reactor-sfr ; Generation IV International Forum, GIF R&D Outlook for Generation IV Nuclear Energy Systems, 31. The SFR can be a breeder reactor.63IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 6. Current designs have an average burnup of around 100 GWd/MT, a power density reaching 300 MW/m3, and a thermal efficiency ranging from 33 to 42 percent.64IAEA, “Advanced Reactors Information System”; Robert Hill, “Sodium Cooled Fast Reactors (SFR)” (presented at GIF Education and Training Task Force Webinar Series 4, December 15, 2016), 27, accessed November 6, 2020, https://www.gen-4.org/gif/upload/docs/application/pdf/2016-12/geniv_sfr_bobhill_final.pdf SFRs can operate in a closed fuel cycle using depleted uranium, MOX fuel, or transuranic waste products from another reactor.65Advanced Reactor Concepts, LLC, “ARC-100: A Sustainable, Cost-Effective Energy Solution for the 21st Century,” last modified 2010, accessed November 6, 2020, https://static1.squarespace.com/static/5b980789a9e0284111acc818/t/5bffefa60ebbe8bc1dd4250c/1543499704783/arc-100-product-brochure.pdf ; Generation IV International Forum, “Sodium-Cooled Fast Reactor (SFR).” SNF mass per unit of energy output is approximately 45 percent that of an LWR.66See Appendix A. Volume per GWye is around 0.129e-3 m3.67See Appendix B. SFR waste can contain sodium from coolant that must be properly treated or separated from other wastes, owing to combustion risk when coming into contact with air and water.68IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 22. There is experimental-scale experience with SFR coolant management, as seen with the Experimental Breeder Reactor-II operational from 1964 to 1994 in the United States.69Nuclear Engineering Division, “Reactors Designed by Argonne National Laboratory,” April 1, 2020, accessed January 14, 2021, https://www.ne.anl.gov/About/reactors/frt.shtml Coolant management will continue to be essential for commercial-scale SFRs.
Lead-cooled Fast Reactor (LFR)
LFRs use liquid lead or lead-bismuth as coolant around solid fuel rods and have an average burnup of 87 GWd/MT.70Based on IAEA, “Advanced Reactors Information System.” Power density is around 100 MW/m3, and thermal efficiency is 40 to 44 percent.71IRSN, Review of Generation IV Nuclear Energy Systems, 163; Craig F. Smith, “Lead-Cooled Fast Reactor (LFR)” (presented at GIF Education and Training Task Force Webinar Series 10, June 12, 2017), 33, accessed November 6, 2020, https://www.gen-4.org/gif/upload/docs/application/pdf/2017-06/geniv-lfr-cfsmith-final.pdf LFRs are intended to operate in closed fuel cycles.72IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 20. The European ALFRED demonstration LFR is planned to use fresh MOX fuel and there is also precedent in French LFRs for recycling MOX fuel.73Generation IV International Forum, GIF R&D Outlook for Generation IV Nuclear Energy Systems, 54; Alessandro Alemberti, “Advanced Lead Fast Reactor European Demonstrator — ALFRED Project,” (presented at GIF Education and Training Task Force Webinar Series 23, September 26, 2018), 14, accessed November 6, 2020, https://www.gen-4.org/gif/upload/docs/application/pdf/2018-11/geniv_alfred_-_alemberti_-final_-_aa.pdf SNF mass per unit of energy output is approximately 46 percent that of an LWR.74See Appendix A. SNF volume per GWye is around 0.9e-3 m3.75See Appendix B. Unique LFR waste primarily takes the form of lead-wetted low- and intermediate-level wastes.76IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 24.
Molten Salt Reactor (MSR)
MSR designs are varied. They can be used as fast or thermal reactors and can process fission products continuously or in batches.77U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors: Nuclear Technology Research and Development, Brian J. Riley et al., (August, 2018): 2, accessed November 6, 2020, https://info.ornl.gov/sites/publications/Files/Pub114284.pdf MSRs can burn actinides, which could reduce radiotoxicity.78IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 1, 6. There are two broad types of MSRs: those that use a molten salt coolant with solid fuel and those that use a liquid fuel-coolant salt.
Solid-fueled MSRs use TRISO particle fuel in molten salt coolants.79U.S. Department of Energy, “TRISO Particles: The Most Robust Nuclear Fuel on Earth,” last modified July 9, 2019, accessed November 6, 2020, https://www.energy.gov/ne/articles/triso-particles-most-robust-nuclear-fuel-earth ; U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 2. The most common solid-fueled MSR is the thermal fluoride-cooled high-temperature reactor (FHR).80IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 7. FHR burnup is approximately two to three times that of LWRs, or 90 to 150 GWd/MT, potentially reaching 190 GWd/MT.81Charles Forsberg and Per F. Peterson, “Spent Nuclear Fuel and Graphite Management for Salt-Cooled Reactors: Storage, Safeguards, and Repository Disposal,” Nuclear Technology 191, no. 2 (2015), 117, https://doi.org/10.13182/NT14-88 ; Jay Disser, Edward Arthur, and Janine Lambert, “Preliminary Safeguards Assessment for the Pebble-Bed Fluoride High-Temperature Reactor (PB-FHR) Concept” (paper presented at the Advances in Nuclear Nonproliferation and Policy Conference, Santa Fe, NM, September 2016), 2, accessed November 6, 2020, https://www.osti.gov/servlets/purl/1358281 Power density is around 10 to 30 MW/m3.82Charles W. Forsberg and Per F. Peterson., “FHR, HTGR, and MSR Pebble-Bed Reactors with Multiple Pebble Sizes for Fuel Management and Coolant Cleanup,” Nuclear Technology 205, no. 5 (2019), 751, https://doi.org/10.1080/00295450.2019.1573619 Thermal efficiency is around 42 percent.83IAEA, “Advanced Reactors Information System.” FHR SNF mass per unit of energy is around 21 percent that of an LWR.84See Appendix A. FHR TRISO fuel pebbles, have a spent fuel volume — around 4.272e-3 m3/GWye — about one-half to one-third that of pebble-fueled VHTRs, with fuel volumes potentially greater than LWRs’ because of the coating and containment material volume.85U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 55; Forsberg and Peterson, “Spent Nuclear Fuel and Graphite Management for Salt-Cooled Reactors,” 114. Like VHTRs, FHRs produce tritium waste at a high rate; Riley et al. (2018) suggest FHR tritium production can be one hundred to one thousand times more than other reactors.86U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 55, 15. Research into tritium management is ongoing; in September 2020 Canadian Nuclear Laboratories and Kairos Power (KP) announced an agreement to research tritium management in the Kairos KP-FHR.87Canadian Nuclear Laboratories, “CNL Partners with Kairos Power on SMR Research,” GlobeNewswire, last modified September 3, 2020, accessed November 6, 2020, https://www.globenewswire.com/news-release/2020/09/03/2088748/0/en/CNL-Partners-With-Kairos-Power-on-SMR-Research.html
Liquid-fueled MSRs continuously circulate and process fuel.88Benjamin R. Betzler, Jeffrey J. Powers, and Andrew Worrall, “Molten Salt Reactor Neutronics and Fuel Cycle Modeling and Simulation with SCALE,” Annals of Nuclear Energy 101:C (2017), 2, https://doi.org/10.1016/j.anucene.2016.11.040; IAEA, “Emerging Technologies Workshop: Trends and Implications for Safeguards” (2017), 24, https://www.iaea.org/sites/default/files/18/09/emerging-technologies-130217.pdf Thermal burnup can range from 29 to 509 GWd/MT and thermal efficiency is around 43 to 46 percent.89IAEA, “Advanced Reactors Information System.” Information on fast MSR burnup is not yet published. Thermal power density is practically around 22 MW/m3.90U.S. Department of Energy, Oak Ridge National Laboratory, Fast Spectrum Molten Salt Reactor Options, David E. Holcomb et al. (July 2011), 5, https://info.ornl.gov/sites/publications/files/Pub29596.pdf Fast MSR power density is higher than that of an LWR, at a practical limit of around 150 MW/m3.91U.S. Department of Energy, Oak Ridge National Laboratory, Fast Spectrum Molten Salt Reactor Options, 4. Thermal spent fuel mass per unit of energy output can be 7 to 125 percent that of an LWR based in part on the wide possible range of MSR burnup values.92See Appendix A. Thermal SNF volume can be around 1.062e-3 m3/GWye.93See Appendix B. Fast MSR SNF mass per unit of energy output is not yet specified, and data are not publicly available to determine approximate SNF volume per GWye.94The approximations in Appendix A rely on a known numerical burnup value. MSRs can use uranium, plutonium, or thorium fuels in a molten fluoride or chloride salt form and can be configured as breeders.95U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors,” 2; Betzler, Powers, and Worrall, “Molten Salt Reactor Neutronics and Fuel Cycle Modeling and Simulation with SCALE,” 23; IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 7. They produce various fission products that require disposal, and can recycle thorium products, uranium produced from thorium fuel, and transuranic materials back into the fuel salt.96Betzler, Powers, and Worrall, “Molten Salt Reactor Neutronics and Fuel Cycle Modeling and Simulation with SCALE,” 19. Significant fission products can be removed to improve fuel efficiency for recirculation.97Betzler, Powers, and Worrall, “Molten Salt Reactor Neutronics and Fuel Cycle Modeling and Simulation with SCALE,” 34. Fluoride salts with lithium produce significant amounts of tritium that, along with fission products and the fluoride itself, contaminate graphite in the core.98David E. Holcomb, “Presentation on Molten Salt Reactor Technology” (presented to US Nuclear Regulatory Commission Staff, Washington, D.C., November 7–8, 2017), 24, accessed November 6, 2020, https://www.nrc.gov/docs/ML1733/ML17331B114.pdf; U.S. Department of Energy, Oak Ridge National Laboratory, Review of Hazards Associated with Molten Salt Reactor Fuel Processing Operations, Joanna McFarlane et al. (2019), 31, https://info.ornl.gov/sites/publications/Files/Pub126864.pdf Liquid-fueled MSRs do not produce irradiated cladding as high-level waste, reducing waste volume from that source.99U.S. Department of Energy, Oak Ridge National Laboratory, Fast Spectrum Molten Salt Reactor Options, 25.
Supercritical Water-cooled Reactor (SCWR)
SCWRs can be thermal, fast, or mixed spectrum, and can be cooled with light or heavy water. Fast SCWRs are significantly less mature than their thermal counterparts, which also require more materials research and development. SCWRs are built on experience with BWRs and serve to increase water-cooled reactor thermal efficiency to around 43 to 48 percent.100Laurence Leung, “Super-Critical Water-Cooled Reactors” (presented at GIF Education and Training Task Force Webinar Series 7, March 28, 2017), 16, accessed October 27, 2020, https://www.gen-4.org/gif/upload/docs/application/pdf/2017-04/geniv_template_laurence_leung_final.pdf The burnups for thermal and fast SCWRs are around 60 GWd/MT and 120 GWd/MT, respectively.101Thomas Schulenberg et al., “Supercritical Water-Cooled Reactor (SCWR) Development through GIF Collaboration” (presented at International Conference onOpportunities and Challenges for Water Cooled Reactors in the 21st Century, Vienna, August 27–30, 2009), 8, accessed November 6, 2020, https://www-pub.iaea.org/MTCD/Publications/PDF/P1500_CD_Web/htm/pdf/topic5/5S06_H.%20Khartabil.pdf SCWRs have similar power densities to LWRs.102IAEA, “Advanced Reactors Information System.” SNF mass per unit of energy output could be around 63 percent that of an LWR for thermal SCWRs or approximately 31 percent for fast SCWRs.103See Appendix A. Thermal SNF volume output could be approximately 25.369e-3 m3/GWye; data to determine fast SNF volume are not available.104See Appendix B. Relevant fast SCWR data are not available for the author to use in approximations. Existing designs are not anticipated to result in wastes significantly different from those of conventional BWRs and represent no significant departures from LWR safeguards.105IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 26, 77; some will be designed as small modular reactors, so any expected waste will be proportional to the size of the reactor as compared to the current fleet; IAEA, “Advanced Reactors Information System”; IRSN, Review of Generation IV Nuclear Energy Systems, 160.
Reactor characteristics for these designs are noted in Figure 2. See Appendix A for more information on the approximated SNF mass per unit of energy output as a percentage of LWRs’ SNF and Appendix B for more information on approximate SNF volume per GWye. See the following section for discussion of the approximations.106Sources used for all types’ characteristics include: U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, hereafter (in this note) Idaho National Laboratory; Buongiorno, “PWR Description,” hereafter in this note Buongiorno; IAEA, “Advanced Reactors Information System”; IAEA, “Other Designs of Nuclear Power Stations,” last modified 2020, accessed October 27, 2020, https://nucleus.iaea.org/sites/graphiteknowledgebase/wiki/Guide_to_Graphite/Other%20Designs%20of%20Nuclear%20Power%20Stations.aspx; U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), hereafter in this note Oak Ridge National Laboratory, VHTR Proliferation Resistance; IRSN, Review of Generation IV Nuclear Energy Systems, hereafter in this note IRSN; Hill, “Sodium Cooled Fast Reactors (SFR), hereafter Hill”; J. Rouault and T.Y.C. Wei., “The GEN IV Gas Cooled Fast Reactor: Status of Studies” (presented at Workshop on Advanced Reactors with Innovative Fuels, Oak Ridge, Tennessee, February 16, 2005), accessed November 6, 2020, https://www.oecd-nea.org/science/meetings/ARWIF2004/2.01.pdf, hereafter Rouault and Wei; Smith, “Lead-Cooled Fast Reactor (LFR),” hereafter Smith; Forsberg and Peterson, “Spent Nuclear Fuel and Graphite Management for Salt-Cooled Reactors,” hereafter Forsberg and Peterson, “Spent Nuclear Fuel and Graphite Management”; Disser, Arthur, and Lambert, Preliminary Safeguards Assessment for the Pebble-Bed Fluoride High-Temperature Reactor (PB-FHR) Concept, hereafter Disser, Arthur, and Lambert; Forsberg and Peterson, “FHR, HTGR, and MSR Pebble-Bed Reactors with Multiple Pebble Sizes for Fuel Management and Coolant Cleanup,” hereafter Forsberg and Peterson, “FHR, HTGR, and MSR Pebble-Bed Reactors”; U.S. Department of Energy, Oak Ridge National Laboratory, Fast Spectrum Molten Salt Reactor Options, hereafter Oak Ridge National Laboratory, Fast Spectrum; Schulenberg et al., “Supercritical Water-Cooled Reactor (SCWR) Development through GIF Collaboration,” hereafter Schulenberg; Leung, “Super-Critical Water-Cooled Reactors,” hereafter Schulenberg et al.; Kyota Uchimura and Akifumi Yamaji, “Preliminary Core Design Study of Small Supercritical Fast Reactor with Single-Pass Cooling,” Journal of Nuclear Engineering 1, no. 1 (2020), https://doi.org/10.3390/jne1010004, hereafter Uchimura and Yamaji. By type, specification sources are as follows: PWR: burnup and thermal efficiency from Idaho National Laboratory, 6–7; power density from Buongiorno, 22; BWR: burnup and thermal efficiency from Idaho National Laboratory, 6–7; power density from IAEA, “Advanced Reactors Information System”; CANDU: burnup and thermal efficiency from Idaho National Laboratory, 7–8; power density from IAEA, “Other Designs of Nuclear Power Stations”; VHTR (prismatic blocks): burnup from Oak Ridge National Laboratory, VHTR Proliferation Resistance, 4; thermal efficiency and power density from IRSN, 51; VHTR (pebble): burnup from Oak Ridge National Laboratory, VHTR Proliferation Resistance, 6; thermal efficiency and power density from IRSN, 51. SFR: Burnup from Hill, 27; thermal efficiency range from IAEA, “Advanced Reactors Information System;” power density from available data in IAEA, “Advanced Reactors Information System” and Hill, 27; GFR: Burnup average and thermal efficiency range from IAEA, “Advanced Reactors Information System”; power density from Rouault and Wei., 9; LFR: Burnup average from IAEA, “Advanced Reactors Information System”; thermal efficiency from Smith, 33; power density average of 84 from IAEA, “Advanced Reactors Information System” and power density of 100 from IRSN, 163; FHR: Burnup from Forsberg and Peterson, “Spent Nuclear Fuel and Graphite Management,” 117, and Disser, Arthur, and Lambert,” 2; thermal efficiency from IAEA, “Advanced Reactors Information System” and supported by Forsberg and Peterson, “Spent Nuclear Fuel and Graphite Management,” 114; power density three to ten times higher than VHTR density, from Forsberg and Peterson, “FHR, HTGR, and MSR Pebble-Bed,” 751; MSR (thermal): Burnup and thermal efficiency range from IAEA, “Advanced Reactors Information System”; power density from Oak Ridge National Laboratory, Fast Spectrum, 5; MSR (fast): Burnup from IRSN, 121; thermal efficiency from Oak Ridge National Laboratory, Fast Spectrum, 4; power density from IAEA, “Advanced Reactors Information System;” SCWR (thermal): Burnup from Schulenberg et al., 8; thermal efficiency from Leung, 16; 73 average power density of from IAEA, “Advanced Reactors Information System,” and power density of 100 from IRSN, 163; SCWR (fast): Burnup from Schulenberg, 8; thermal efficiency from Uchimura and Yamaji, 47; power density from Schulenberg et al., 3.
Fig. 2: Reactor Characteristics.
Nuclear waste, and SNF in particular, has been a concern associated with nuclear power since its inception. With few national plans for SNF management implemented, countries will face compounded difficulties from emerging reactors in addition to the pre-existing challenges of managing a global sum of over 400,000 metric tons of heavy metal of conventional reactor SNF.107Stimson Center, “Spent Nuclear Fuel Storage and Disposal.” The differences between emerging waste forms and volumes and those of the current fleet result in new challenges that require consideration throughout the design and development process. Three back-end challenges associated with innovative fuels, coolants, and fuel cycles are: 1) bridging the gap between new waste features and existing final disposal strategies, 2) squaring recycling-based designs with national policies or practices prohibiting reprocessing, and 3) identifying ways to adapt international safeguards to the reactors’ fuel outputs.
Challenge 1: Treatment and Final Disposal
Disposal of SNF in a DGR has been considered the best option for long-term management.108OECD Nuclear Energy Agency, “Management and Disposal of High-Level Radioactive Waste: Global Progress and Solutions,” Timothy McCartin (2020), 9, accessed November 6, 2020, https://www.oecd-nea.org/jcms/pl_32567/management-and-disposal-of-high-level-radioactive-waste-global-progress-and-solutions First proposed in the 1950s, deep geological disposal has been determined through 40 years of research to effectively reduce safety risks to people and the environment and limit potential acquisition of radioactive material for malicious purposes.109OECD Nuclear Energy Agency, “Progress Towards Geological Disposal of Radioactive Waste: Where Do We Stand?: An International Assessment” (1999), 7, 9, 11, accessed November 6, 2020, https://www.oecd-nea.org/jcms/pl_13268 ; OECD Nuclear Energy Agency, “Management and Disposal of High-Level Radioactive Waste,” 17. After a thorough geographical siting assessment and safety review process, a DGR would be constructed and then be operational — accepting waste canister emplacement in its tunnels — for as many as one hundred years.110OECD Nuclear Energy Agency, “Management and Disposal of High-Level Radioactive Waste,” 25. After that time, it would be sealed to contain the waste. Research is also ongoing into alternative geological disposal methods, including deep vertical or horizontal borehole drilling, rather than tunneling; these methods could change disposal siting and have different interactions with SNF decay heat.111Stimson Center, “Spent Nuclear Fuel Storage and Disposal”; comments on disposal conceptualization changes from Pavel Hejzlar, in interview with the author, October 26, 2020. While waste management is recognized as a national responsibility, that responsibility does not preclude safe management in a multinational geological repository, where multiple countries would coordinate a waste disposal repository.112In 2003, IAEA Director-General Mohamed ElBaradei noted the value of considering “multinational approaches to the management and disposal of spent fuel and radioactive waste” as a result of the large number of extant temporary storage sites, restrictive geological requirements for DGRs, and potential financial and resource burden of a national DGR for some states, from Mohamed ElBaradei, “Statement to the Fifty-Eighth Regular Session of the United Nations General Assembly,” IAEA, last modified November 3, 2003, accessed November 6, 2020, https://www.iaea.org/newscenter/statements/statement-fifty-eighth-regular-session-united-nations-general-assembly; the Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management can be considered to have proposed the possibility of MGR collaboration: IAEA, “Developing Multinational Radioactive Waste Repositories: Infrastructural Framework and Scenarios of Cooperation,” (2004), 1–2, https://www.iaea.org/publications/7135/developing-multinational-radioactive-waste-repositories-infrastructural-framework-and-scenarios-of-cooperation For more information on safeguards and management implications for multinational geological repositories, see Stimson Center, “Back-end to the Future: Some Safeguards Considerations for Multinational Geological Repositories,” Cindy Vestergaard and James Casterton, last modified January 2020, accessed November 6, 2020, https://www.stimson.org/2020/back-end-to-the-future-some-safeguards-considerations-for-multinational-geological-repositories/ Some private entities, for example Deep Isolation in the United States, are also beginning to look at how the private sector can contribute to long-term waste management responsibilities.113Deep Isolation, “Nuclear Waste Disposal Solutions,” accessed January 1, 2021, https://www.deepisolation.com/ Along with DGR construction underway in Finland, with fuel emplacement expected this decade, research is ongoing into other repositories in 11 countries.114Finland’s timeline from Posiva Oy, “General Time Schedule for Final Disposal,” accessed October 29, 2020, http://www.posiva.fi/en/final_disposal/general_time_schedule_for_final_disposal#.X5sEoYhKgdU ; these countries are France, Sweden, Canada, China, Czech Republic, Germany, India, Japan, Russia, Switzerland, and the United Kingdom. Canada, Nuclear Waste Management Organization (Canada), “Programs Around the World for Managing Used Nuclear Fuel,” last modified 2018, accessed October 23, 2020, https://www.nwmo.ca/~/media/Site/Files/PDFs/2018/04/09/09/55/Programs-Around-the-World-2018_web.ashx?la=en With both operational disposal facilities and advanced reactors on the horizon, future commercialization will affect planned repositories, particularly as DGR decisions continue to be delayed, as they are in the United States.115U.S. Department of Energy, Blue Ribbon Commission on America’s Nuclear Future, Report to the Secretary of Energy, Lee H. Hamilton et al. (2012), vi, accessed January 26, 2021, https://www.energy.gov/sites/prod/files/2013/04/f0/brc_finalreport_jan2012.pdf The previous section highlighted some of the primary SNF differences between emerging reactors and LWRs. Here, the resultant challenges will be explored in more depth, including through comparing SNF mass and volume between different reactor types. These challenges are largely possible to overcome with enough attention paid to finding technical solutions. Stakeholders should consider the emerging differences in back-end management as disposal plans and research into unconventional reactors’ commercialization develop.
Fuel output volume is likely to be significantly affected as emerging reactors become incorporated into fleets (See Figure 3). TRISO fuel for use in VHTRs and FHRs complicates waste management because of its unusual bulky form, which will necessitate different packing for spent fuel storage and disposal. Combined with a volume output per GWye comparable to or potentially greater than that of an LWR, as may be the case for the prismatic VHTR, significant volume increases may occur over a unit’s lifetime. Conversely, reactor types that can breed fuel or produce energy over longer periods without refueling — for example some fast reactors or MSRs — may have significantly smaller waste volumes. Disposal volume considerations will eventually depend on the balance between next-generation reactors’ small plant size and the size of advanced reactor fleets to guarantee energy output.
Fig. 3: SNF Volume per GWye by Reactor Type.116See Appendix B for more information on volume approximations and discussion of possible sources of error.
Disposal volume considerations will eventually depend on the balance between next generation reactors’ small plant size and the size of advanced reactor fleets to guarantee energy output.
Mass of spent fuel also differs between reactor types. Generally, SNF mass per unit of energy output ranges from about seven percent that of an LWR, seen in an MSR with a burnup of about 500 GWd/MT, to 154 percent that of an LWR, seen with an example SFR.117See Figure 2 in the section, “Emerging Reactor Classes and Their SNF”; for methodological explanation, see Appendix A. Most emerging reactors including some MSRs, FHRs, LFRs, GFRs, and VHTRs, will produce a smaller mass of SNF per unit of energy output than LWRs.118For more discussion of possible sources of error, see Appendix A. Like volume, the total impact on storage and disposal facilities, as well as waste management practices is dependent on the number of reactors deployed to reach target energy outputs. For example, one Kairos Power 140 MWe KP-FHR would produce about 20 percent the amount of SNF of an 1100 MWe pressurized water reactor (PWR). However, if enough KP-FHRs are operated to produce the same output as that PWR, then the total SNF mass per unit of energy increases approximately 160 percent.119See Appendix A for calculations. KP-FHR information from Kairos Power, “Technology,” last modified 2020, accessed November 6, 2020, https://kairospower.com/technology/ This balance will need to be carefully quantified to ensure adequate capacity at all back-end facilities.
Repository design changes can also stem from emerging reactor specifications. TRISO SNF’s durability allows storage at higher temperatures than LWR fuel rods. If TRISO fuel becomes popular, it could deepen disposal facilities, if other factors allow.120U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 69. According to the IAEA, salt and deep borehole disposal are preferable to overcome liquid-fueled MSRs’ higher heat, narrowing disposal options for this type of SNF.121U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 69. Of the three DGRs for SNF around the world where site selection has already occurred, none are in salt formations, and of the nine countries currently evaluating potential SNF/HLW DGR sites, only Germany is currently considering siting in salt.122Nuclear Waste Management Organization (Canada), “Programs Around the World for Managing Used Nuclear Fuel,” 3. Deep borehole disposal is being researched as a quicker, cheaper, and more isolated approach than tunneling.123For example, research into deep borehole technology being done at the University of Sheffield: The University of Sheffield, “Nuclear Engineering,” 2021, accessed January 26, 2021, https://www.sheffield.ac.uk/materials/research/themes/nuclear-engineering#geo With higher decay heat than LWRs, MSR waste canisters would likely be smaller and potentially compatible with the typical borehole limit of 20 inches diameter.124U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 69; Stimson Center, “Evolving Technologies for Future Deep Geological Repositories: A Closer Look,” Trinh Le (2020), accessed November 6, 2020, https://www.stimson.org/2020/evolving-technologies-for-future-deep-geological-repositories-a-closer-look/ However, research into fuel salt waste forms is ongoing, meaning waste volume and canister dimensions are not yet certain. Countries lacking suitable geography or other factors in support of a salt or deep borehole disposal will benefit from thinking about future advanced reactor deployment in reverse, starting from planned (or eventually, existing) repositories, in order to ensure that commercial-scale forays into reactors such as liquid-fueled MSRs are compatible with available disposal practices.
Advanced fuels will require additional treatment or innovative containment methods before final disposal. TRISO pebbles are likely to be directly disposed of and used only in open — or once-through — cycles, given the difficulty of separating the SNF from the outer coating for reprocessing, and the IAEA suggests that “[t]he coatings are ideal for a multiple barrier waste management system.”125IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 107. For prismatic TRISO blocks, removal of fuel components from the blocks for disposal is also under consideration to reduce volume.126Fukaya and Nishihara, “Reduction on High Level Radioactive Waste Volume and Geological Repository Footprint with High Burn-Up and High Thermal Efficiency of HTGR,” 192. Without coating separation, new canister development is likely necessary to account for fuel element geometry and volume. MSR salt is highly corrosive and additional treatment is necessary to remove residue from solid fuel elements.127Forsberg and Peterson, “Spent Nuclear Fuel and Graphite Management for Salt-Cooled Reactors,” 115. MSR liquid salt fuel will require encapsulation and final disposal strategies different from those for existing fuel rod disposal. Research is ongoing into glass, ceramic-metal composite, and other methods of immobilizing salt fuel for disposal, with the primary obstacle being salt corrosion.128U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 36. In addition, used graphite moderators from liquid MSRs will contain residual fuel salt amounts from the liquid, making it more difficult to separate SNF from other HLW without further processing prior to disposal.129U.S. Department of Energy, Pacific Northwest National Laboratory and Oak Ridge National Laboratory, Identification of Potential Waste Processing and Waste Form Options for Molten Salt Reactors, 53. There are also challenges associated with SFR coolant as a result of liquid sodium’s reactivity to water, requiring specific treatment conditions after removal from a reactor.130Columbia University Center on Global Energy Policy, A Comparison of Advanced Nuclear Technologies, 67. Lead coolant toxicity and corrosiveness also requires additional processing, based on existing methods, before disposal.131IAEA, Waste from Innovative Types of Reactors and Fuel Cycles, 24.
An additional treatment under investigation around the world is the partitioning and (sometimes) transmutation (P&T) of MAs — particularly neptunium, curium, and americium — from SNF and potential recycling in fast reactors to reduce overall SNF radiotoxicity and decay heat output.132United Kingdom, National Nuclear Laboratory, “Minor Actinide Transmutation: Position Paper” (2014), 4, accessed November 6, 2020, https://www.nnl.co.uk/wp-content/uploads/2019/01/minor_actinide_transmutation_-_position_paper_-_final_for_web1.pdf. P&T could reduce overall waste radiotoxicity one hundred to one thousand times (although multiple studies have suggested that peak radiological dose rates would be unaffected) and could increase DGR capacity by a factor of 2 to 50, depending on time of emplacement after cooling.133United Kingdom, National Nuclear Laboratory, “Minor Actinide Transmutation: Position Paper,” 4; S. David and S. Massara, “Impact on Nuclear Scenarios with Gen IV and ADSs” (presented at the OECD-NEA Second International Workshop on Technology and Components for Accelerator-Driven Systems, Nantes, France, May 2013), 14, https://www.oecd-nea.org/science/wpfc/tcads/2nd/presentations/documents/0.03-TCADS-SdavidSMassaraV2.pdf; Serco, “Literature Review of Partitioning and Transmutation,” N. Butler (2011), 4, https://www.ensterna.com/wp-content/uploads/2017/04/Literature-review-of-partitioning-and-transmutation.pdf; United Kingdom, National Nuclear Laboratory, “Minor Actinide Transmutation,” 9. From 2002 to 2007, the IAEA conducted a coordinated research project culminating in a report on Advanced Reactor Technology Options for Utilization and Transmutation of Actinides in Spent Nuclear Fuel.134IAEA, Advanced Reactor Technology Options for Utilization and Transmutation of Actinides in Spent Nuclear Fuel, IAEA-TECDOC-1626 (2010), https://www.iaea.org/publications/8214/advanced-reactor-technology-options-for-utilization-and-transmutation-of-actinides-in-spent-nuclear-fuel Related fast reactor-based projects around the world continue to advance, such as the Belgian MYRRHA project, which experiments with a proton-driven subcritical LFR.135IAEA, Advanced Reactor Technology Options for Utilization and Transmutation of Actinides in Spent Nuclear Fuel (2010); MYRRHA, “MYRRHA Project,” last modified 2020, accessed November 9, 2020, https://www.myrrha.be/myrrha-project/ Advanced thermal reactors have also been proposed — for example pebble bed VHTRs with larger uranium fuel pebbles and smaller burnable MA pebbles.136Forsberg and Peterson, “FHR, HTGR, and MSR Pebble-Bed Reactors with Multiple Pebble Sizes for Fuel Management and Coolant Cleanup,” 750, 751. Research is ongoing into more experimental laser pulse-accelerated transmutation methods as well.137Scott Birch, “High-Power SYLOS Laser Could ‘Transmute’ Nuclear Waste,” Reuters Events: Nuclear, Aug 5, 2019, accessed November 9, 2020, https://analysis.nuclearenergyinsider.com/high-power-sylos-laser-could-transmute-nuclear-waste The prospective effectiveness of P&T for each MA varies.138United Kingdom, National Nuclear Laboratory, “Minor Actinide Transmutation,” 5. As P&T requires separation of radioactive material, there is also the concern that separated MAs could be targets for theft and use as radiological weapons.139United Kingdom, National Nuclear Laboratory, “Minor Actinide Transmutation,” 12. Notably, the French Institute for Radiological Protection and Nuclear Safety (IRSN) reports that transmutation would lead to increased transportation flows — assuming the continuing use of existing or similar transportation packages — leading in turn to potential proliferation concerns.140IRSN, Review of Generation IV Nuclear Energy Systems, 218. As research into technology feasibility continues, nonproliferation considerations will be necessary for those states that choose to incorporate P&T into advanced fuel cycles.
Small reactors (under 300 MWe output) of all types will also impact spent fuel transportation to final disposal sites, and new fuels will impact how SNF transportation occurs. As developers including Ultra Safe Nuclear Corporation have noted, small reactors can be sited in remote areas with smaller power grids.141Ultra Safe Nuclear Corporation, “MMR™ Energy System,” accessed January 14, 2021, https://usnc.com/mmr-energy-system/ While more remote locations and smaller facilities can reduce plant and material access, increasing control over some potential proliferation pathways, there will also be more reactor facilities in disparate locations where there previously were none, introducing new attractive material to would-be bad actors. These isolated plants may require longer routes to centralized storage or disposal facilities, which increases proliferation risk because material in transit is vulnerable to interception and could travel along many different routes.142IAEA, “Transport Security,” last modified 2020, accessed October 21, 2020, https://www.iaea.org/topics/transport-security
In addition, TRISO or MSR liquid fuel as well as some designs’ sealed cores, transported in one piece to factories for refueling, will require engineering and other technical transport considerations. Rodney McCullum, senior director of Fuel and Decommissioning at the Nuclear Energy Institute, notes that transportation of new fuel forms is primarily an engineering question: “because these are dual purpose systems, you have to be able to take the storage system and put it on a railcar or on a truck . . . you don’t want to package these things up and then repackage them when it comes time to ship them to a repository.”143Rodney McCullum, interview with the author, September 28, 2020. To address this challenge, timely technical focus will be essential. In addition, rather than having fuel rods lined up in a container, TRISO pebbles or blocks will have to be oriented differently. Given the larger volume of TRISO SNF compared to fuel rods, with 300–400 pebble additions, and therefore removals, each day, transporting TRISO fuel could be costlier and require more shipments, depending on the eventual size of each container.144United States, Idaho National Engineering and Environmental Laboratory, Advanced Core Design And Fuel Management For Pebble Bed Reactors, Hans D. Gougar, Abderrafi M. Ougouag, and William K. Terry (October 2004), 18, accessed October 21, 2020, https://inldigitallibrary.inl.gov/sites/sti/sti/3310868.pdf Based on SNF canister specifications from the Chinese HTR-PM pebble-bed VHTR, a canister for TRISO could hold a pebble volume of 4.52 m3, or 40,000 “spent fuels” (Appendix C).145See Appendix C. Zhang Zuoyi et al., “Future Development of Modular HTGR in China after HTR-PM” (presented at the 7th International Topical Meeting on High Temperature Reactor Technology, Weihai, China, October 28, 2014), 5, accessed November 6, 2020, https://nucleus.iaea.org/sites/htgr-kb/HTR2014/Paper%20list/Track1/HTR2014-11456.pdf Those 40,000 pebbles are approximately 10 percent of the core loading.146Zhang et al., “Future Development of Modular HTGR in China after HTR-PM,” 4, 5. The U.S. Standardized Transportation and Disposal canister can hold four PWR assemblies, or about two percent of a typical core.147Buongiorno, “PWR Description,” 11. The waste canister to be used in Finland’s DGR can also hold four PWR assemblies.148Svensk Kärnbränslehantering Aktiebolag, Spent Nuclear Fuel for Disposal in the KBS-3 Repository, Technical Report TR-10-13, Per Grahn, Lena Morén and Marie Wiborgh (December 2010), 37, accessed January 28, 2021, https://www.osti.gov/etdeweb/servlets/purl/1030179 As with repository volume, with a larger number of plants needed to achieve the same energy output, pebble transport volume can rapidly overtake that of LWRs.
The above challenges related to spent fuel volume and post-reactor SNF treatment are by no means insurmountable and are largely technical. For many, research into potential technical solutions is ongoing. However, these technical challenges are set against a backdrop of continued debate around spent fuel management, which means disposal pursuits are only now accelerating. As Pavel Hejzlar, technical fellow at TerraPower, notes about back-end challenges, “The greatest challenge for us and other reactor developers is that there is no permanent repository.”149Pavel Hejzlar, interview with the author, October 26, 2020. It is therefore very important to consider the impact of emerging reactors on the larger back-end and disposal ecosystem.
Challenge 2: Reprocessing/Recycling
Another means of waste management is the reprocessing, and potential subsequent recycling, of spent fuel for easier disposal or reuse in a reactor. Conventional PUREX hydrometallurgical reprocessing separates plutonium and uranium from the other components in SNF, creating MOX fuel or reprocessed uranium fuel.150IAEA, Spent Fuel Reprocessing Options, IAEA-TECDOC-CD-1587 (2009), 13, accessed November 6, 2020, https://www.iaea.org/publications/8143/spent-fuel-reprocessing-options This process is shown in Figure 4.
The advantage of reprocessing is that, regardless of reactor type, it inherently decreases the volume of SNF headed for final disposal, as it allows about 96 percent of the material in SNF to be reformed into fresh fuel.152This does not mean that the entirety of the 96 percent of SNF is depleted in its second cycle, but reuse allows more fission products to be removed from the material headed for disposal. Belfer Center, The Economics of Reprocessing vs. Direct Disposal of Spent Nuclear Fuel, Matthew Bunn et al. (2003), 3, https://www.belfercenter.org/publication/economics-reprocessing-vs-direct-disposal-spent-nuclear-fuel By reducing SNF quantity, the process makes SNF storage facilities less attractive as sources for nuclear material theft or diversion and offers potential reassurance about disposal site safety. Reusing fuel is not unique to advanced non-light-water systems; for example, France has reprocessed plutonium from SNF into MOX fuel for use in several PWRs since the 1960s.153Shant Krikorian, “France’s Efficiency in the Nuclear Fuel Cycle: What Can ‘Oui’ Learn,” IAEA, last modified September 4, 2019, accessed November 6, 2020, https://www.iaea.org/newscenter/news/frances-efficiency-in-the-nuclear-fuel-cycle-what-can-oui-learn However, cost and proliferation concerns have resulted in movement away from or prohibition of full reprocessing in national practices and official policies. In 2007, the Director of the U.S. Congressional Budget Office testified to the Senate Committee on Energy and Natural Resources that the cost of reprocessing in the U.S. would be 25 percent more than the cost of direct disposal.154Peter R. Orszag, Testimony on Costs of Reprocessing Versus Directly Disposing of Spent Nuclear Fuel before the Committee on Energy and Natural Resources, United States Senate (November 14, 2007), available from Congressional Budget Office, 3, https://www.cbo.gov/sites/default/files/cbofiles/ftpdocs/88xx/doc8808/11-14-nuclearfuel.pdf
Proliferation concerns about conventional reprocessing arise because pure plutonium is separated, making it an attractive material for diversion towards weapons. Because of proliferation concerns, Sweden’s SNF practice shifted in the 1970s from reprocessing to preparation for direct disposal, engendering the current DGR project.155Sweden, Ministry of the Environment, Sweden’s Sixth National Report Under the Joint Convention on the Safety of Spent Fuel Management and on the Safety of Radioactive Waste Management: Sweden’s Implementation of the Obligations of the Joint Convention, Björn Dverstorp et al. (2017), 20, https://www.iaea.org/sites/default/files/sweden-nr-6th-rm-jc.pdf South Korea’s 1974 civil nuclear cooperation agreement with the United States requires a once-through domestic fuel cycle; however, the South Korea–U.S. Joint Fuel Cycle Study exploring pyro-processing and other waste management possibilities has been scrutinized as a departure from that condition.156U.S. Congressional Research Service, CRS Insights: U.S.-Republic of Korea Nuclear Cooperation Agreement, Mary Beth D. Nikitin and Mark Holt (2015), accessed January 20, 2021, 1, https://fas.org/sgp/crs/nuke/IN10304.pdf ; Robert Einhorn, “U.S.-ROK Civil Nuclear Cooperation Agreement: Overcoming the Impasse,” Brookings, last modified October 11, 2013, accessed November 6, 2020, https://www.brookings.edu/on-the-record/u-s-rok-civil-nuclear-cooperation-agreement-overcoming-the-impasse/ Currently, only France, Russia, and India commercially operate SNF reprocessing facilities — Japan’s Rokkasho Reprocessing Plant sits unused and the UK’s Magnox Reprocessing Plant has recently begun its permanent closure process.157World Nuclear Association, “Processing of Used Nuclear Fuel,” last modified June 2019, accessed November 6, 2020, https://www.world-nuclear.org/information-library/nuclear-fuel-cycle/fuel-recycling/processing-of-used-nuclear-fuel.aspx ; World Nuclear News, “Sellafield Starts Controlled Shutdown of Magnox Facility,” last modified March 23, 2020, accessed November 6, 2020, https://world-nuclear-news.org/Articles/Sellafield-starts-controlled-shutdown-of-Magnox-fa The United States did operate a facility from 1966 to 1972 for military fuel reprocessing, but it was closed when regulations were heightened.158U.S. Congressional Research Service, Nuclear Fuel Reprocessing: U.S. Policy Development, Anthony Andrews (March 27, 2008), CRS-2, http://large.stanford.edu/courses/2014/ph241/parekh2/docs/RS22542.pdf
Despite these concerns regarding conventional reprocessing, many emerging reactor designs, for example the already-operational Russian BN-800 SFR, incorporate fuel processing and recycling either as a fundamental function or possible configuration for more sustainable fuel use.159World Nuclear News, “First Serial Batch of MOX Fuel Loaded into BN-800,” last modified January 28, 2020, accessed November 6, 2020, https://world-nuclear-news.org/Articles/First-serial-batch-of-MOX-fuel-loaded-into-BN-800 GFRs, LFRs, SFRs, and fast MSRs and SCWRs can be configured to use recycled SNF to operate in alternative fuel cycles. Developers are eager to differentiate between the innovative processing facilities they envision for their fuel cycles and existing large-scale reprocessing plants. Everett Redmond, senior technical advisor at the Nuclear Energy Institute, draws a distinction between conventional reprocessing and modern techniques for some emerging reactors: “Fast reactors are a different energy spectrum and so you have the ability to operate and deal with the fuel in a different form, so those companies aren’t looking at something like La Hague.”160La Hague is a reprocessing plant in France. Quote from Everett Redmond, interview with the author, September 28, 2020. While some proposed techniques will differ from conventional reprocessing, it is likely that developers will need to continue clarifying the difference between newer processing proposals and conventional reprocessing to assuage proliferation concerns.
To provide a compelling argument for investment and deployment, developers must reassure stakeholders that their designs are not only cost-effective, but also inherently equipped with unprecedented levels of safety and proliferation controls, including in waste management. Several notable reactor designs emphasize their proliferation risk reduction resulting from the ability to recycle fuels from LWRs and the depleted uranium from fuel fabrication, while not fully separating plutonium from other materials. One such reactor is the ARC-100 SFR, under pre-licensing vendor design review in Canada.161ARC Energy, LLC, “Background,” accessed January 14, 2021, https://www.arcenergy.co/ Designs that rely on recycling fuel as a primary waste management solution or nonproliferation benefit may encounter political reluctance to closed fuel cycles. Even for countries that do not expressly prohibit such operations, most lack reprocessing infrastructure, making international movement of spent fuel between reactors, temporary storage facilities, and reprocessing facilities more likely and potentially opening up fuel cycles to more proliferation pathways. Policy resistance to closed cycles will likely continue past the time of initial deployment, in which case developers whose designs can operate in closed cycles must consider other ways in which their reactors will provide innovations in waste management and nonproliferation, or risk falling behind other once-through reactors.
Policy resistance to closed cycles will likely continue past the time of initial deployment, in which case developers whose designs can operate in closed cycles must consider other ways in which their reactors will provide innovations in waste management and nonproliferation, or risk falling behind other once-through reactors.
This concern may seem premature considering that most designs are still in the conceptual or pre-licensing phase, but there is already limited commercial deployment of these advanced technologies, as seen in Russia. While several of the countries with significantly advanced development in the space — China, Russia, India — allow reprocessing, countries that are not explicitly planning on reprocessing are also researching closed-cycle advanced reactors, such as South Korea with its PEACER LFR under conceptual design.162IAEA, “Advanced Reactors Information System.” In addition, by incorporating future plans for partially or fully closed fuel cycles into their designs, developers are beginning to highlight the reprocessing discussion, even if the deployment ramp for reprocessing or recycling facilities is still long. In the U.S., Third Way Senior Policy Advisor Jackie Kempfer notes that “[m]any of the advanced reactors in the later stages of development do not require reprocessing, so this is not an issue that needs to be fully resolved before we see the first U.S. reactors deployed.”163Jackie Kempfer, email to the author, October 22, 2020.
The degree to which emerging reactors will inspire policy change in reprocessing or recycling remains to be seen. With current policy reluctance partially tempering the speed of innovative reprocessing deployment, there is currently an opportunity to clarify and discuss future reprocessing and recycling intentions and develop alternative nonproliferation approaches.
Challenge 3: Back-End Safeguards
In addition to safe SNF management, developers hoping to deploy their designs will have to consider how to apply international safeguards to material and facilities, especially since several countries leading advanced reactor development or interested in these emerging reactors have obligations to safeguard material under comprehensive safeguards agreements (CSAs) and additional protocols (APs) with the IAEA.164There are several non-safeguarded advanced reactor projects globally. For example, the U.S. Department of Defense’s Project Pele aims to develop a microreactor for siting at forward-deployed military location. From U.S. Department of Defense, “DOD Awards Contracts for Development of Mobile Microreactor,” last modified March 9, 2020, accessed January 20, 2021, https://www.defense.gov/Newsroom/Releases/Release/Article/2105863/dod-awards-contracts-for-development-of-a-mobile-microreactor/. Even for designs undergoing development in nuclear-weapon States with voluntary offer agreements rather than CSAs, safeguards application is critical for any future export as the Nuclear Suppliers Group guidelines mandate CSAs for non-nuclear-weapon States importing nuclear material or technology.165IAEA, “Communication Received from the Permanent Mission of the Czech Republic to the International Atomic Energy Agency Regarding Certain Member States’ Guidelines for the Export of Nuclear Material, Equipment and Technology (INFCIRC/254/Rev.12/Part 1)” (2013), 1, https://www.iaea.org/sites/default/files/publications/documents/infcircs/1978/infcirc254r12p1.pdf Many emerging reactors’ fuel and core designs have safeguards advantages built in — for example in the inherent difficulty of fully separating plutonium from molten fuel. These features generally make up “Safeguards by Design” (SBD). The IAEA notes that built-in safeguards can lessen the burden of verification in the back-end: “The successful maintenance of CoK [Continuity of Knowledge] to reduce the need to reverify spent fuel requires facility features. . . ”166IAEA, International Safeguards in the Design of Facilities for Long Term Spent Fuel Management, Nuclear Energy Series No. NF-T-3.1, Brian Boyer, James Sprinkle, and Gary Dyck (2018), 15, https://www.iaea.org/publications/10806/international-safeguards-in-the-design-of-facilities-for-long-term-spent-fuel-management However, SBD alone is not sufficient to meet international obligations under CSAs and APs; applied safeguards monitoring and verification are still necessary, particularly for new reactors. One safeguards expert noted that, in contrast to the safeguards process for conventional reactors, SBD is really “safeguards by early discussion and participation, where previously we had the luxury of implementing model approaches in familiar facilities once the legal requirements to provide information arose.”167Interview with the author, September 22, 2020.
Accordingly, while inherent features do provide safeguards and nonproliferation benefits, ongoing and early interaction on safeguardability is necessary to ensure that safeguards activities as a whole are cost-efficient, robust, and harmonious with back-end management operations. Early engagement will help to create networks between developers, national regulators and the IAEA, providing sufficient time for the IAEA to adequately prepare and implement for safeguards approaches before deployment.
While inherent features do provide safeguards and nonproliferation benefits, ongoing and early interaction on safeguardability is necessary to ensure that safeguards activities as a whole are cost efficient, robust, and harmonious with back-end management operations.
One safeguardability challenge is related to the novel fuels that emerging reactors introduce to the fuel cycle. For example, the IAEA has noted that at DGRs, reconceptualization of material accounting to account for bulk TRISO will be necessary in material balance areas prior to canister placement.168A material balance area is defined as “an area in or outside of a facility such that: (a) The quantity of nuclear material in each transfer into or out of each ‘material balance area’ can be determined; and (b) The physical inventory of nuclear material in each ‘material balance area’ can be determined when necessary, in accordance with specified procedures, in order that the material balance for Agency safeguards purposes can be established”; IAEA, IAEA Safeguards Glossary: 2001 Edition, International Nuclear Verification Series No. 3 (2002), 46–47, accessed January 28, 2021, https://www.iaea.org/publications/6663/iaea-safeguards-glossary; note about DGR reconceptualization based on the existing safeguards plan for the Finnish DGR: Posiva Oy, Design of the Disposal Facility 2012, Timo Saanio et al. (2013), 126–127, https://inis.iaea.org/search/search.aspx?orig_q=RN:45087772 The current fleet has itemized fuel, specifically fuel elements that can be individually counted.169IAEA, International Safeguards in the Design of Fuel Fabrication Plants, Nuclear Energy Series No. NF-T-4.7, Brian Boyer, James Sprinkle, and Gary Dyck (2017), 39, 26, accessed November 6, 2020, https://www.iaea.org/publications/10746/international-safeguards-in-the-design-of-fuel-fabrication-plants TRISO pebbles will be treated as bulk material, which is defined by the IAEA as “material in loose form, such as liquid, gas, or powder, or in a large number of small units (e.g. pellets or pebbles) that are not each individually identified for material accountancy purposes.”170IAEA, IAEA Safeguards Glossary, 34. Examples of bulk handling facilities are fuel fabrication or conversion facilities; there is no operational precedent for treating a reactor and associated storage like a bulk handling facility, and there are no current proposals for bulk TRISO disposal.171IAEA, International Safeguards in the Design of Fuel Fabrication Plants, 5. Adapting this procedure for a reactor facility will require innovative accountancy methods instead of tracking individual fuel elements based on serial number, especially after removal from the reactor.172U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), 5. One possibility is the incorporation of bulk “near real time accountancy,” in which inventory data are provided by the operator to the IAEA nearly immediately, rather than waiting for annual operator inventory accounting.173IAEA, IAEA Safeguards Glossary, 46. Since a pebble-bed reactor is refueled and defueled online and could at any one time contain upwards of 200,000 fuel pebbles circulating, enhanced monitoring beyond fuel rod–based core methods is necessary.174There are to be about 200,000 pebbles, in the Xe-100 reactor under Vendor Design Review in Canada and in development under the U.S. Advanced Reactor Demonstration Program: X-Energy, “Reactor: Xe-100,” last modified 2020, accessed November 6, 2020, https://x-energy.com/reactors/xe-100 Some guidance on other potential safeguards measures already exists, including monitoring fuel flow to determine radiation levels for each pebble.175IAEA, International Safeguards in the Design of Nuclear Reactors, Nuclear Energy Series No. NP-T-2.9, James Sprinkle, Donald Kovacic, and Matthew van Sickle (2014), 26, accessed November 6, 2020, https://www.iaea.org/publications/10710/international-safeguards-in-the-design-of-nuclear-reactors
The challenges for pebbles lie in ensuring robust CoK throughout and after their use, expanding monitoring of the greater number of fuel elements than a conventionally fueled reactor has, and preventing diversion during frequent SNF handling as a result of online refueling.176IAEA, International Safeguards in the Design of Nuclear Reactors, 25. James Casterton, a nonresident fellow at the Stimson Center with experience in safeguards work at the Canadian Nuclear Safety Commission and the IAEA, notes that “when you’re dealing with different wastes, for example some that are liquid, some that are spherical, it will require probably as an initial step, [a change in] the way in which encapsulation is being considered at this point in time.”177Interview with the author, September 24, 2020. A potential change in encapsulation methods creates complexity by increasing the number of process streams at a facility, or increasing the number of such facilities to separate the different processes. In addition, the IAEA notes, bulk facilities typically require more intensive verification than item facilities, so these changes could pose inspection capacity and cost-effectiveness challenges if bulk-fueled reactors are to become prevalent.178IAEA, International Safeguards in the Design of Fuel Fabrication Plants, 5. This concern necessitates further discussion to effectively integrate bulk encapsulation safeguards into planned disposal facilities.
MSR liquid fuel, also considered bulk, will require even more advanced monitoring and accounting methods than TRISO pebbles, which represent a dramatic change for reactor and fuel accounting but can still reasonably incorporate bulk accountancy.179University of Maryland Center for International & Security Studies, “Safeguards-By-Design for Advanced Nuclear Systems,” Lance K. Kim (2017), 5, https://drum.lib.umd.edu/handle/1903/19699; IAEA, International Safeguards in the Design of Nuclear Reactors, 29. Bulk accountancy instruments from facilities at other parts of the fuel cycle cannot be directly applied to liquid MSRs because, as Andrew Worrall, section head of the Integrated Fuel Cycle Section at Oak Ridge National Laboratory, points out, “[the fuel is] always flowing, not in the same place, isotopically changing, and incredibly intensely radiologically and temperature-hot.”180Vienna Center for Disarmament and Non-Proliferation, “The Impact of Small Modular Reactors on Nuclear Nonproliferation and IAEA Safeguards,”Nicole Virgili (2020), 32, https://vcdnp.org/the-impact-of-smrs-on-non-proliferation-and-iaea-safeguards/; Andrew Worrall’s comment from interview with the author, October 2, 2020. Moreover, material accountancy in the back-end is as yet unclear for MSRs, as a final disposal form for the liquid fuel has not been determined.
The monitoring and inspection of LFRs, MSRs, and SFRs face obstacles from corrosiveness, toxicity, and, for LFRs and SFRs, coolant opacity.181U.S. Congressional Research Service, Advanced Nuclear Reactors: Technology Overview and Current Issues, Danielle A. Arostegui and Mark Holt (April 18, 2019), 22, https://crsreports.congress.gov/product/pdf/R/R45706 These concerns also exist in handling the fuel and waste after removal from reactors, requiring more research into materials and monitoring strategies. The IAEA has suggested joint use of operators’ remote viewing systems during inspection as an option to better inspect areas with opaque coolant.182IAEA, International Safeguards in the Design of Nuclear Reactors, 24. In MSRs specifically, the fission products still partially incorporated into the fuel during and after operation disrupt non-destructive assay (NDA) readings.183IAEA, “Emerging Technologies Workshop: Trends and Implications for Safeguards,” 24. Advanced monitoring methods to take advantage of the differences between MSRs and conventional reactors are still being researched.
Fast reactors configured to breed material from irradiated fuel pose safeguards risks because of their plutonium production, and their potentially heightened inspection needs necessitate a discussion of fast reactor impact on safeguards optimization. The U.S. Integral Fast Reactor project (1984–1994) developed a design where reactor operation, reprocessing without plutonium separation, and recycling could all be done at one site, decreasing the need for monitoring at multiple sites and aiming to decrease proliferation pathways.184U.S. Department of Energy, Argonne National Laboratory, “Reactors Designed by Argonne National Laboratory: Integral Fast Reactor,”last modified November 8, 2017, accessed November 6, 2020, https://www.ne.anl.gov/About/reactors/integral-fast-reactor.shtml However, waning political interest in nuclear power ended the program, and such infrastructure is still in early development elsewhere, as it is based on unconventional types of processing.185PBS, “Dr. Charles Till | Interviews | Frontline,” last modified 1996, accessed November 6, 2020, https://www.pbs.org/wgbh/pages/frontline/shows/reaction/interviews/till.html ; Columbia Center on Global Energy Policy, A Comparison of Nuclear Technologies, 67. Collocating these types of facilities would reduce proliferation pathways from transit, but would likely necessitate more intrusive safeguards and safeguards at a greater number of sites.186IAEA, International Safeguards in the Design of Nuclear Reactors, 29. Existing NDA procedures could be applied to fast reactor spent fuel, but research into safeguarding fuel with significant amounts of MAs, which can disrupt NDAs, is ongoing.187IAEA, International Safeguards in the Design of Nuclear Reactors, 29. The IAEA has noted that fast reactor facilities will likely be inspected more frequently and with more measurements than thermal reactors.188The FBRs planned in India will not be safeguarded: Belfer Center, “India’s Nuclear Safeguards: Not Fit for Purpose,” John Carlson (2018), 4, https://www.belfercenter.org/sites/default/files/files/publication/India%E2%80%99s%20Nuclear%20Safeguards%20-%20Not%20Fit%20for%20Purpose.pdf; IAEA, International Safeguards in the Design of Nuclear Reactors, 29; Yukiya Amano, “Challenges in Nuclear Verification,” IAEA, last modified April 5, 2019, accessed November 6, 2020, https://www.iaea.org/newscenter/statements/challenges-in-nuclear-verification This consideration calls safeguards inspection capacity into question in the face of developers’ eagerness to pursue widespread deployment; more frequent inspections at more, isolated facilities could strain IAEA inspector capacity and safeguards optimization. In 2019, then-IAEA Director General Yukiya Amano said that while from 2010 to 2018 the number of safeguarded facilities globally rose by 12 percent, the number of significant quantities of safeguarded material rose by 24 percent, and the number of material accountancy reports rose by over 33 percent, the IAEA’s safeguards budget during that same period increased by only 6.3 percent (real terms).189Amano, “Challenges in Nuclear Verification.” He noted that despite a concerted effort to expand remote monitoring and surveillance, he anticipated increased inspector strain moving forward.190Amano, “Challenges in Nuclear Verification.” This change occurred over a decade of light water–cooled thermal reactor additions; incorporating more fast reactors with diverse fuel types is likely to exacerbate the challenge. Early coordination among developers, national bodies, and the IAEA on this front would help to prepare and train inspectors for new demands.
The IAEA has noted that “if fuel movements are performed without human access and access to fuel storage locations are similarly limited, remote monitoring of the fuel movements by reliable, redundant systems can reduce the need for on-site inspections.”191IAEA, International Safeguards in the Design of Nuclear Reactors, 28. Monitoring innovative reactor SNF will likely rely significantly on remote monitoring, as a result of the technical and capacities challenges addressed above. On the other hand, more frequent emplacement of a wider variety of fuels would need more intensive verification and a larger variety of safeguards methods.192IAEA, International Safeguards in the Design of Facilities for Long Term Spent Fuel Management, 36.
The challenges outlined in this section are primarily related to ensuring that safeguards are considered in the design process and simultaneously exploring new or more flexible safeguards approaches, all the while evaluating the impacts on inspector capacity. It is beneficial for all parties to initiate these considerations early to avoid retrofitting safeguards elements and to ensure safeguards requirements are not relaxed. One expert in the field noted, “Even if it is harder to continue to rely on material accountancy in new designs and with new kinds of fuels . . . it is important to try to preserve that standard of safeguards rather than decrease verification capabilities by accepting less rigorous standards.”193Interview with the author, September 16, 2020. Even for designs being developed in nuclear-weapon States, stakeholders benefit from furthering robust approaches to SBD and safeguards applications, particularly to ensure a sturdy safeguards foundation for future deployment in non-nuclear-weapon States. With direct engagement among relevant parties, practical and effective solutions can be found before safeguards activities need to be implemented.
Work is accelerating around the world to support the eventual deployment of an emerging generation of advanced reactors that will impact the nuclear fuel cycle. As they develop, comprehensive back-end management should be just as foundational for design development as front-end or in-reactor features. Countries have not yet implemented final management strategies for the conventional fuel accumulating in storage facilities around the world. With new reactors employing block, pebble, and liquid fuels, and a host of commercially unprecedented coolants, care must be taken to ensure that challenges in their management are addressed to limit the burden they place on existing waste management. As these disruptive technologies develop, it will be tempting to delay decisions on complicated emerging reactor waste, as existing SNF inventory concerns have not yet been resolved. Instead, this dramatic shift in nuclear energy is an opportunity to work in earnest on existing SNF concerns.
Stakeholders need to communicate and decide whether recycling, transmutation, or other strategies will be implemented, and how emerging non-LWRs will be incorporated into existing or planned disposal strategies. Policy around reprocessing is not likely to change before these reactors’ deployment, but policy reluctance means that reactor developers cannot rely solely on reprocessing and recycling for nonproliferation and SNF management and must consider additional nonproliferation layers. The IAEA must be adequately prepared to take on a fleet of new reactors requiring innovative monitoring, verification, and inspection, necessitating a synthesis between intrinsic safeguards and institutional approaches. It is possible to overcome the challenges new SNF and fuel cycles pose, but not if management strategies and safeguards are addressed in the eleventh hour. The advent of a new generation of power reactors is a pivotal opportunity for all stakeholders to address concerns about SNF.
As these disruptive technologies develop, it will be tempting to delay decisions on complicated emerging reactor waste, as existing SNF inventory concerns have not yet been resolved. Instead, this dramatic shift in nuclear energy is an opportunity to work in earnest on existing SNF concerns.
Appendix A: SNF Mass Per Unit of Energy Output
Based on typical reactor class specifications, it is possible to approximate SNF mass per unit of energy output for new reactor classes as percentages of existing LWRs’ SNF mass output. These results are not intended to be authoritative calculations, but rather general approximations of the comparative SNF mass per unit of energy output. An Idaho National Laboratory assessment uses enrichment to calculate SNF mass output.194U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, A-1 – A-3. Based on the assumption that burnup and enrichment are related through energy output, burnup can also be used.195Andrew Favalli et al., “Determining Initial Enrichment, Burnup, and Cooling Time of Pressurized-Water-Reactor Spent Fuel Assemblies by Analyzing Passive Gamma Spectra Measured at the Clab Interim-Fuel Storage Facility in Sweden,” Nuclear Instruments and Methods in Physics Research Section A: Accelerators, Spectrometers, Detectors and Associated Equipment 820 (2016), 8, https://doi.org/10.1016/j.nima.2016.02.072 Based on the Idaho research, the equation used here to approximate SNF mass per unit of energy output is shown as follows:
where unit capacity factor (UCF) is expressed as a percentage and represents the actual output a plant produces divided by possible energy output in a year;196IAEA, “Glossary of Terms in PRIS Reports,” 2021, accessed January 18, 2021, https://pris.iaea.org/PRIS/Glossary.aspx thermal efficiency is expressed as a percentage and represents how efficiently a reactor turns heat into electrical energy;197Collins Dictionary, “Thermal Efficiency,” 2021, accessed January 27, 2021, https://www.collinsdictionary.com/us/dictionary/english/thermal-efficiency burnup represents the rate at which uranium (or fissile material) is used up in the reactor and is expressed in gigawatt-days per metric ton (GWd/MT).198U.S. Nuclear Regulatory Commission, “Backgrounder on High Burnup Spent Nuclear Fuel,” (2018), accessed January 27, 2021, https://www.nrc.gov/reading-rm/doc-collections/fact-sheets/bg-high-burnup-spent-fuel.html
To find new reactors’ percentage of SNF mass per unit of energy output as compared to LWRs’, the time of operation is held constant between new reactors and LWRs to find the SNF mass per unit of energy output rather than over operating lifetimes or other periods. The final equation used is shown as a ratio of the new reactor’s SNF output to the LWR’s:
Relevant data values, and calculated SNF mass per unit of energy output for all calculated reactors are shown below. Most UCFs are calculated as the average of listed reactors in the IAEA Advanced Reactors Information System (ARIS).199IAEA, “Advanced Reactors Information System”; this approach is informed by the methodology used by Idaho National Laboratory, “Nuclear Fuel Cycle Evaluation and Screening — Final Report,” Roald Wigeland et al. (2014), 9, accessed January 19, 2021, https://fuelcycleevaluation.inl.gov/Shared%20Documents/ES%20Main%20Report.pdf to provide more comprehensive data rather than relying on a specific technology. When there are not enough data for an average, example reactor specifications from that database are used. For the MSR, approximations can better be expressed by showing multiple calculations for wide burnup ranges.
|Reactor type||Typical discharge burnup (GWd/MT)||Typical thermal efficiency (%)||Typical/ expected UCF (%)||SNF mass per unit of energy output as a percentage of LWR SNF mass/unit energy output||SNF mass per unit of energy output (kg)|
|LWR200LWR data from U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 6, 7.||50||33||90||100||54.55|
|CANDU201U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 7, 8.||7||29.3||90||804.49||438.81|
|VHTR (prismatic)202Burnup from U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), 4; thermal efficiency from IRSN, Review of Generation IV Nuclear Energy Systems, 51; UCF from average of available data in IAEA, “Advanced Reactors Information System.”||120||45||91.5||31.06||16.94|
|VHTR (pebble bed)203Burnup from U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), 6; thermal efficiency from IRSN, Review of Generation IV Nuclear Energy Systems, 51; UCF from HTR-PM specifications in IAEA, “Advanced Reactors Information System.”||90||45||85||38.48||20.99|
|GFR204All information based on average of available data in IAEA, “Advanced Reactors Information System.”||121||43||92.5||32.95||17.97|
|SFR205Burnup from IRSN, Review of Generation IV Nuclear Energy Systems, 163; thermal efficiency is average of available values in IAEA, “Advanced Reactors Information System”; UCF from 4S specifications in the IAEA, “Advanced Reactors Information System.”||100||38.9||95||44.77||24.42|
|LFR206Specifications from average of available data in IAEA, “Advanced Reactors Information System.”||87||42.4||92.7||46.07||25.13|
|FHR207Burnup and thermal efficiency from specifications for Mk-1 PH-FHR in IAEA, “Advanced Reactors Information System”; UCF from average of 5 listed reactors in IAEA, “Advanced Reactors Information System.”||180||42.5||90||21.57||11.77|
|Thermal MSR (low-range burnup)208Burnup is listed for the Integral Molten Salt Reactor in IAEA, “Advanced Reactors Information System”; thermal efficiency is the average of MSRs with listed thermal efficiencies in IAEA, “Advanced Reactors Information System”; UCF is the average of all values listed in IAEA, “Advanced Reactors Information System.”||29||44.6||88.5||125.44||68.42|
|Thermal MSR (high-range burnup)209High-range MSR burnup and thermal efficiency comes from the ThorCon reactor in the IAEA, “Advanced Reactors Information System”; UCF is the average of all values listed in the IAEA, “Advanced Reactors Information System.”||509||46.4||88.5||6.87||3.75|
|SCWR (thermal)210Discharge burnup and thermal efficiency are the average of available values in IAEA, “Advanced Reactors Information System.” In the absence of published expected UCF, SCWR UCF is assumed to be 90 percent, which is typical for existing reactors.||60||43.7||90||62.93||34.33|
|SCWR (fast)211Burnup from Schulenberg et al., “Supercritical Water-Cooled Reactor (SCWR) Development through GIF Collaboration,” 8; thermal efficiency from Uchimura and Yamaji, “Preliminary Core Design Study of Small Supercritical Fast Reactor with Single-Pass Cooling,” 47.||120||44||90||31.25||17.05|
Possible sources of error arise from the inconsistency across reactor types of the specifications available. Where possible, specifications are an average of available specifications in the IAEA ARIS, if typical values (see Fig. 2, “Reactor Characteristics”) represent a range. Because the system relies on vendor reporting to inform its database and the format differences of those reports provided, many reactors in the database do not have the necessary specifications listed to precisely find type-based SNF mass per unit of energy output. In addition, the database does not include all designs (as some designers have not provided information). In the case of the SFR, the only design that had the UCF available was an outlier in other typical specifications. In that case, more representative specifications were used. The pebble-bed VHTR and SFR estimations are based on the only UCFs available in ARIS; as such, these figures can be considered examples rather than general approximations. Where no UCF information was available, a UCF of 90 percent, typical for existing reactors, was assumed.212U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 6. The thermal MSR was assigned both a high-range burnup specification and low-range burnup because of the wide range of values, while the fast MSR was not included because of lack of available data on burnup. This set of estimations is intended to be a starting point for consideration moving forward, rather than an authoritative set of data about these reactors, given the high likelihood of error based on data availability and consistency.
For the Kairos Power example:
A 140 MWe FHR, like the KP-FHR, produces approximately 20 percent of an 1100 MWe LWR’s SNF mass output per unit of energy. Reaching the same energy output with FHRs in a region previously receiving energy from an 1100MWe LWR would require approximately 8 FHRs.
Therefore, total SNF mass in the region for equivalent energy output would be
or approximately 160 percent of current LWR-based mass.
Appendix B: Spent Fuel Volume Produced per GWye
The following are comparative approximations of spent fuel output volume per GW year of electricity produced for the different reactor classes with available relevant data. These approximations do not account for SNF canister capacities, canister packing fractions or other external considerations that would affect SNF volume as it enters a disposal or storage facility. Instead, the following approximations provide general comparative estimates for the volume of fuel units themselves immediately after removal from a reactor. Data are drawn primarily from reactor specifications within the ARIS database. The average of all available values for each reactor type is taken, with significant outliers excluded, to provide a more general example of expected fuel volume.
Volume produced is based on the volume of a single fuel unit (for example, a fuel assembly, bundle, or pebble, for which volume is calculated based on active fuel height and pitch or diameter, depending on the geometry), the number of units in the core, and the refueling cycle and amount removed during each refueling period. To find volume per GWye, a 1 GW per year plant is assumed. For MSRs, which do not have solid fuel units, total core fuel salt volume is used. The following approximations account for outages, when more SNF is not being produced and fuel could be removed (for offline-refueling reactors), by incorporating expected or typical UCF.213U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, A-2. For reactors with continuous or online refueling, UCF can provide information about the amount of fuel units being discharged over a time frame, as more spent fuel is not being produced when the reactor is offline for maintenance or other reasons.
The volume of spent fuel produced per GWye can be approximated with the following:
|Reactor Type||Fuel Unit||Avg. or typical fuel unit width/ pitch/ diameter (mm)||Avg. active fuel height (mm)||Single Fuel Unit Volume (m3)||Fuel Units in Core||Refueling Cycle (y)||Fraction of Core Removed During Refueling||Avg. UCF (%)||SNF Volume (m3/ GWye)|
|PWR214Assembly dimensions and quantity from Buongiorno, “PWR Description,” 22; refueling period and core fraction from U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 6.||Square Assembly||215||3660||0.16918||193||1.5||0.33||90||6.530e-3|
|BWR215Assembly dimensions from Jacopo Buongiorno, “BWR Description,” 9; number of assemblies from Buongiorno, “BWR Description,” 22; refueling period and core fraction from U.S. Department of Energy, Idaho National Laboratory, Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 6.||Square Assembly||137.54||3707.9||0.07014||764||1.5||0.33||90||10.718e-3|
|CANDU216Fuel unit volume based on bundle diameter and active fuel height, Nuclear Waste Management|
Organization, “What iIs Used Nuclear Fuel?,” 2016, accessed January 15, 2021, 2, https://www.nwmo.ca/~/media/Site/Files/PDFs/2016/11/10/12/38/EN_Backgrounder_UsedNuclearFuel_LowRes.ashx?la=en#:~:text=Each%20CANDU%20fuel%20bundle%20is,mass%20of%20about%2024%20kilograms; bundle number in core from Atomic Energy of Canada Limited, “Characteristics of Used CANDU Fuel Relevant to the Canadian Nuclear Fuel Waste Management Program,” K. M. Wasywich (1993), 20, https://www.osti.gov/etdeweb/servlets/purl/169800; refueling cycle and core fraction from U.S. Department of Energy, Idaho National Laboratory., Plutonium Discharge Rates and Spent Nuclear Fuel Inventory Estimates for Nuclear Reactors Worldwide, 7, 8.
|VHTR (prismatic)217Specifications from average of available entries in IAEA, “Advanced Reactors Information System”||Hexagonal Prismatic Block||381.33||881||0.11094||800||1.66||0.5||91.5||24.461e-3|
|VHTR (pebble-bed)218Sphere diameter from U.S. Department of Energy, Oak Ridge National Laboratory, Very High-Temperature Reactor (VHTR) Proliferation Resistance and Physical Protection (PR&PP), 23; other specifications from average of available values in IAEA, “Advanced Reactors Information System.”||Spherical Pebble||60||N/A||0.00011||390,000||continuous||350/day or 0.000897 of core/day||85||12.281e-3|
|GFR219Specifications from average of available values in IAEA, “Advanced Reactors Information System.” Specifications for the ALLEGRO reactor in IAEA, “Advanced Reactors Information System” supported by Ladislav Bělovský, “The ALLEGRO Experimental Gas Cooled Fast Reactor Project” (presented at GIF Education and Training Task Force Webinar Series 27, March 20, 2019), 17, 19, 20, accessed January 18, 2021, https://www.gen-4.org/gif/upload/docs/application/pdf/2019-03/geniv_template-_dr._ladislav_belovsky_final_3-20-19.pdf||Hexagonal Assembly||115.75||2280||0.02204||415||10.93||0.7||92.5||0.774e-3|
|LFR220Specifications taken as averages of available values in IAEA, “Advanced Reactors Information System.”||Hexagonal Assembly||280||883.3||0.05998||151.8||9.38||1||92.7||0.9e-3|
|SFR221SFR specifications from the average of available values in IAEA, “Advanced Reactors Information System,” excluding the 30-year refueling cycle outlier from the Toshiba 4S system; fraction of core removed during refueling from Tanju Sofu, “SFR Technology Overview” (presented at Fast Reactor Technology Training, U.S. Nuclear Regulatory Commission, March 26, 2019), 21, accessed January 18, 2021, https://www.nrc.gov/docs/ML1914/ML19149A378.pdf||Hexagonal Assembly||72.2||1212.5||0.00547||121.75||1.64||0.33||95||0.129e-3|
|FHR222Fuel pebble diameter and quantity are from MIT, “Technical Description of the ‘Mark 1’ Pebble-Bed Fluoride Salt-Cooled High-Temperature Reactor (PB-FHR) Power Plant,” Charalampos Andreades et al. (2014), 38, https://web.mit.edu/nse/pdf/researchstaff/forsberg/FHR%20Point%20Design%2014-002%20UCB.pdf; fraction removed daily from MIT, “Technical Description of the ‘Mark 1’ Pebble-Bed Fluoride Salt-Cooled High-Temperature Reactor (PB-FHR) Power Plant,” 114, 115.||Spherical Pebble||30||N/A||0.00001||470,000||continuous||920/day or 0.00196 of core/ day||90||4.272e-3|
|MSR (thermal)223Volume from ThorCon entry, IAEA, “Advanced Reactors Information System”; additional specifications from average of available data in IAEA, “Advanced Reactors Information System.”||Liquid Fuel Salt||N/A||N/A||8.4||1||14.667||1||88.5||1.062e-3|
|SCWR (thermal)224Specifications from average of available values in IAEA, “Advanced Reactors Information System.”||Square Assembly||286||4200||0.163||645.67||1.09||0.29||90||25.369e-3|
These volume approximations should be considered baseline examples for general comparative analysis rather than authoritative figures because of the likelihood of error from available data inconsistency across types and the prospective nature of data for technologies not yet implemented. Emerging reactor designs are at varying conceptual development stages and are intended for different purposes, making precise calculation difficult. The above approximations address these differences by considering average values across reactor types. Notably, UCF values are from the average of available values from ARIS, as seen in Appendix A. If UCF data for a given reactor type is unavailable, a UCF of 90, typical for existing reactors, is used (see Appendix A). Using the average of expected values, as well as the current typical industry average, helps to account for potential bias in reported specifications, while still acknowledging that features differing from current commercial specifications, including prospective UCF, will have different impacts on back-end outputs.225As in Appendix A, this approach is informed by U.S. Department of Energy, Idaho National Laboratory, Nuclear Fuel Cycle Evaluation and Screening — Final Report, 9, to provide more comprehensive data about a reactor type instead of individual expectations for technologies.
MSRs are an unusual case, as the constant circulation of fuel will mean that volume output per GWye is highly dependent on the details of the design and the amount of time the core operates, with fuel being discharged only at the end of the core life.226As seen in the plans for the Integral Molten Salt Reactor-400 from IAEA, “Advanced Reactors Information System.” The fast SCWR and fast MSR are not included because of lack of available data.
Appendix C: Transport Containers
According to HTR-PM reactor specifications, the TRISO pebble diameter is 60mm.227Zhang et al., “Future Development of Modular HTGR in China after HTR-PM,” 4–5.
Volume is then 1.13×10-4 m3. The HTR-PM reactor specifications note that there are 40,000 “spent fuels,” or pebbles, in a waste canister, making total SNF volume in a waste canister = 4.52 m3.228Zhang et al., “Future Development of Modular HTGR in China after HTR-PM,” 5.
The HTR-PM core holds approximately 400,000 pebbles, meaning a waste canister can hold 10 percent of the total core loading.229Zhang et al., “Future Development of Modular HTGR in China after HTR-PM,” 4.
In comparison, the U.S. STAD (Standardized Transportation and Disposal) canister, holds 4 PWR fuel rods, which is approximately two percent of the total core loading in a 17 ´ 17 fuel assembly array.230Buongiorno, “PWR Description,” 14.
Appendix D: Glossary of Terms
|AGR||Advanced Gas-cooled Reactor|
|AP||Additional Protocol (to a Comprehensive Safeguards Agreement)|
|ARIS||Advanced Reactors Information System (from the International Atomic Energy Agency)|
|BWR||Boiling Water Reactor|
|CANDU||Canadian Deuterium Uranium (type of pressurized heavy water reactor)|
|CoK||Continuity of Knowledge (Safeguards)|
|CSA||Comprehensive Safeguards Agreement|
|DGR||Deep Geological Repository|
|FBR||Fast Breeder Reactor|
|FHR||Fluoride-cooled High Temperature Reactor|
|GFR||Gas-cooled Fast Reactor|
|GWd/MT||Gigawatt-days per metric ton (unit of fuel burnup rate)|
|GWye||Gigawatt-year of electrical power (unit of electrical production)|
|HTGR||High Temperature Gas-cooled Reactor|
|IAEA||International Atomic Energy Agency|
|LFR||Lead-cooled Fast Reactor|
|LILW||Low- and Intermediate-level Waste|
|LWR||Light Water Reactor|
|MOX||Mixed Oxide (fuel)|
|MSR||Molten Salt Reactor|
|MTHM||Metric Tons of Heavy Metal|
|MW/m3||Megawatts per cubic meter (unit of power density)|
|MWe||Megawatts of electrical output (unit of electrical production)|
|NPT||Treaty on the Non-Proliferation of Nuclear Weapons|
|P&T||Partitioning and Transmutation|
|PHWR||Pressurized Heavy Water Reactor|
|PWR||Pressurized Water Reactor|
|SBD||Safeguards by Design|
|SCWR||Supercritical Water-cooled Reactor|
|SFR||Sodium-cooled Fast Reactor|
|SNF||Spent Nuclear Fuel|
|TRISO||Tristructural Isotropic (fuel)|
|UCF||Unit Capacity Factor|
|VHTR||Very High Temperature Reactor|
This research was made possible by the generous support of the John D. and Catherine T. MacArthur Foundation. The author is grateful to the Stimson Center Nuclear Safeguards Program’s Director Cindy Vestergaard and Research Analyst Trinh Le for their guidance and assistance in developing this product. The author also thanks all parties who were willing to offer their insights for inclusion in the paper and those who conducted peer review.